/
1 FINANCIAL MARKETS, ACCOUNTING & MANAGEMENT 1 FINANCIAL MARKETS, ACCOUNTING & MANAGEMENT

1 FINANCIAL MARKETS, ACCOUNTING & MANAGEMENT - PowerPoint Presentation

luanne-stotts
luanne-stotts . @luanne-stotts
Follow
355 views
Uploaded On 2018-11-16

1 FINANCIAL MARKETS, ACCOUNTING & MANAGEMENT - PPT Presentation

ACADEMIC YEAR 20172018 DOTT DOMENICO DALLOLIO MAIL domenicodalloliouniveit Skype profddo 2 PART 1 INVESTMENTS how to take effective investing decisions in the algo trading era ID: 729694

buy price market time price buy time market order risk book trading shares average sell money auction step trend

Share:

Link:

Embed:

Download Presentation from below link

Download Presentation The PPT/PDF document "1 FINANCIAL MARKETS, ACCOUNTING & MA..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

Slide1

1

FINANCIAL MARKETS, ACCOUNTING & MANAGEMENT

ACADEMIC YEAR

2017-2018

DOTT. DOMENICO DALL’OLIO

MAIL:

domenico.dallolio@unive.it

Skype: prof_ddoSlide2

2

PART 1

INVESTMENTS:

how to take effective investing decisions in the algo trading eraSlide3

3

3

What is an investment?

An investment is the commitment of money and/or other resources

today

under the expectation to get some kind of benefit

tomorrow

.

It is not all just about the

money

: sometimes we invest our

time

in activities given of some

expected future value

, possibly higher than the actual one.

You are here today because you are investing your time and your money in

exchange

for skills and competences you expect will give you the chance to fully achieve your professional and personal goals.

The main risk here is that the real value of an investment could be much different from the expected value!

A common feature in any investment: we sacrifice something today

under the hope

to get something of a higher value

tomorrow. Only in the end we will know if we did right or not.

It’s a

sacrifice

because in allocating money to an investment we give up the chance to use that money for something

else, and we cannot know which choice would be the best,

until it will be too late.

It’s the so called

opportunity cost of capital

.Slide4

4

4

What is an investment?

The problem is due to the fact our capital is limited: given some amount of money, time or other resources we always have to choose within a range of investment alternatives, all theoretically replaceable one another.

In the end we will have to pick only one or some, and this puts us in front of an implied cost, given by the risk that after a while we could find out that our choices were not optimal.

An issue here is

how much choices are replaceable one another

: to be able to state that choices A and B are replaceable implies that we are able to compare them perfectly, so to be able to state which one is better (or that they are completely repleaceble each other). But since we cannot exactly foretell which of the two will actually be better in the end, we can only take decisions in

uncertainty conditions

and

incomplete information

.

What we need in the first place is to find a unique

yardstick

, some way to compare different investment choices and to be able to sort them on some scale, that is to decide which one(s) provide better

expected

results.Slide5

5

A paramount concept

In a few words, in

the field of

investments

it is all just about risk and return

In making an investment usually people ask themselves how much they could gain; very rearly the expectation about profits is based on scientific criteria: more often it is the result of a hope originated by the will to satisfy a contingent need of some sort.

Only a few investors take time to think better about this central issue and ask themselves the only right question:

how much could I reasonably earn given the risk I am assuming

?

The strict relationship between risk and return is the starting key-point: it is not only a matter of where I am in the beginning and where I am in the end, but also of

how I get from the beginning to the end

.

The path of an investment in time, indeed, is as much important as its final result.Slide6

6

It is all just about risk and return

The basic rule, then, is that return always comes with risk:

where there is no risk there cannot be any return

.

Experience in the field teaches that under

“normal” market conditions

(that is, when there are no external factors coming into play) risk

and return are substantially proportional: if

we want to gain more, we have to suffer a higher level of risk. And if risk increases we want the chance to gain more, otherwise we will not take that risk.

Here a paramount problem comes into play, though: it is all a matter of

perceived risk

and

expected return:

real

levels of risk and return are hardly predictable in the moment

we take the investment. Indeed,

investors can only make

hypothesis, in the first place.

Real results will be revealed only

at the end of the

period

.

Here

psychological issues

also come into play: given the impossibility to take decisions on the basis of sure facts, tolerable risk levels are different among different investors. Generally speaking, different investors acting on the markets tend to have different risk tolerance levels, and different expectations in terms of returns, even when they are put in front of the same investment opportunities.

Risk, by the way,

is subjective

: ask ten different investors to quantify risk of an investment and you will likely get ten different evaluations!Slide7

7

Misleading advertising

Today surfers on the internet are pounded over and over by banners and advertising spaces pointing out at easy profits on the

financial markets, usually on the Forex market or on binary options or on CFDs. The story is more or less always the same: it takes just a few euros to open the account, 15 minutes to read a pdf and understand everything, 20 minutes a day to earn 5k euros per week.

It is misleading advertising under several points of view:

with just a few euros on the account you will survive (financially speaking) only a few minutes: the leverage that allows you to trade with just a few euros can make you gain a lot but also lose a lot;

return is not possible without risk, and to gain a lor of money you need a lot of money (apart from very rare exceptions);

the trading job is as sophisticated as brain surgery; if it takes 10 years to become a good brain surgeon, how could it possibly take only a few minutes to become a successful trader?

The point is not to be able to gain a thousand euros in one day,

but to be able to do it every day, to end up after a year with a positive balance on the account, having suffered a bearable risk in between.

Once again, the path of the equity line is as important as its final value.Slide8

8

A small step ahead

One of the most dangerous mistakes many people do is to think they can make a lot of money without taking any risk and with a very little capital. The truth is that the trader who makes a 25% return

systematically

every year is an excellent trader. Moreover if his returns have two characteristics:

low volatility among different years (stable profits year after year);

low volatility within each year: to have a +25% at the end of the year after being -50% at some time during that year is not positive.

In order to understand the importance of volatility among different years, let’s consider two investments: one makes a 20% return for three consecutive years, the other makes +50% the first year, -30% the second, +40% the third. They have the same final result (not considering compounding), but for sure they are not equal!

Stable earnings, almost predictable, in the first case; totally unpredictable in the second.

For the third time: the path of an investment is as much important as its final value.Slide9

9

A small step ahead

Speaking of the volatiity within each year, the problem is to be able to define a sustainable threshold: some people cannot stand the risk of losing some part of their capital; some other don’t mind.

The fact is exactly this:

the risk tolerance level is not the same for all investors

.

And moreover, given the same perceived risk, different investors could likely have different expectations about a return sufficient to justify taking that risk.

There’s a clear subjectivity issue in the evaluation of risk and return.

Market models suggest that

two investments having the same perceived risk level should have the same expected return

, otherwise not a single investor would ever take the one with a lower expected return.

The fact is that

this statement is based on some very strong assumptions

, as we’ll see very soon

.Slide10

10

And vice-versa: two investments given of the same expected return should expose the investors to the same level of risk.

Being all of this true, investors could put all investments on a scale of risk and return and be always able to pick the one that best fits their risk tolerance level and expectations about returns.

In

the real world this

often doesn’t

happen, given the

asymmetries in the availability of information

, that make hard for many investors to evaluate all the possible scenarios and to compare all the alternative investments.

There are some open problems then:

how can we define and measure risk?

how can we make expectations about return?

can we build models to solve this kind of issues?

do these models work?

A small step aheadSlide11

11

Looking back to the two investments in slide

8,

the main problem is the

volatility

(usually

measured

with the standard

deviation of past returns)

of results in the two cases: in the first there is no volatility, since the value of the return is always the same

. This makes future returns quite predictable.

Should it have been +22% the first year, +18% the second, +20% the

third,

things would have only be slightly different: in fact, you could say that for the fourth year you could expect a +20%, being prepared to face something more or less between +18% and +22%.

In the other case it is very hard to say that you could expect a 20% for the fourth year: it is exactly the mean of the yearly past returns, but to say that it should be the return for the fourth year appears clearly hazardous. This because the volatility of returns is very high, and this means we have to deal with a

strong unpredictability

.

And here comes a paramount statement:

volatility is a synonim of unpredictability

.

A small step aheadSlide12

12

Why volatility is so important

We can then state a fundamental

rule:

the

mean of past returns

can be taken as an expectation of future returns only

when the volatility

(of past returns)

is not too high.

Indeed,

volatility

i

s a measure of the reliability of an expected return

.

Statistics applied to finance can

then tell

us something about the

reasonable expected return

of an investment, and the

probability

of that return: in simple words,

volatility is a measure of

risk

to make a mistake in our forecasts about returns

.

This probability can be determined under a

model

(that’s what modern financial engineering does), or after the

observation

of historical data series.Slide13

13

The Normal model

Should it be possible to state that returns of any financial asset follow a Normal model, it would be very easy to measure the exact probability of any interval of values, given the mean and the standard deviation of historical

returns.

This way we could form a good expectation about the return of an investment and the risk to miss that expectation,

but

this wouldn’t solve a major issue: indeed it would provide a measure for the

(expected) risk

of mistake,

but not a way to reduce it.

Volatility is crucial in the definition of the risk of any investment, since depending on its value the risk to make a mistake can be significant.

But the key-issue here is

that

the volatility we know in the moment we take the investment is the past one, while the one we will have to deal with is the

future one, expected but unknown.Slide14

14

The higher the volatility, the higher the dispersion of returns: the shape of the distribution gets squashed in the middle and the tails rise. In these situations it is said that the distribution has

fat tails

.

When volatility increases, forecasts become very hard and unreliable, since the field of variation of possible results gets much wider than usual, and events usually given of very low odds to show up become much more probable.

Extreme events, the so called black swans, become more probable.

As we’ll see, by the way, the Normal model is just an approximation of reality:

the real world is much riskier than the model!

How do models deal with risk and return?

Why is the Normal important in finance?Slide15

15

Portfolio theory by Harry Markovitz

The portfolio theory by Harry Markovitz states that the risk of investments can be reduced investing in portfolios of lowly correlated assets: since stocks and other assets tend to move in different ways most of the time we can assume that in any observed timeframe some assets would move up, some other down, some other would stay stable. In one word,

diversification

.

Return

would then be the weighted average of the returns of all the assets included in the portfolio, while

global

risk

would be reduced thanks to the low correlation between pairs, that would smooth the risk of each single component taken on its own.

Some issues here:

historical returns are not a good forecast of future ones;

historical volatility is not a good measure of future one;

correlations between pairs are not stable in time;

the effect of diversification is also a function of weights of the single components, which can be optimized only on past data; and once again, future optimal weights could be different from past ones.Slide16

16

CAPM

The capital asset pricing model, by the way, states that the return of an asset or a portfolio can be seen as a linear combination of risk free interest rates, market risk premium and risk of the asset (or the portfolio). Basicly, the idea is that

riskier assets should always provide higher returns

.

Some issues here too:

the linear relationship between risk and return is a function of active and passive interest rates, assumed to be equal;

risk of an asset or a portfolio is an expectation based on past data;

market risk premium is based on the expected return of the market portfolio, which is a weighted average of returns expected on each component;

risk of any asset is measured by its Beta, which is a ratio between a correlation and a variance, both expected;

the model implies that investors always have a full information and can therefore take the best decisions in terms of risk and return

In three words:

too many forecasts

!Slide17

17

A different point of view

What follows is an approach to

investments

completely different from the one that comes out from models like the CAPM.

The idea to start from is that not all the investors on a market have access to the same information, furthermore they do not all have the same skills, and moreover they are not always rational. It is then possible to find investing strategies able to provide some kind of edge in respect with other inverstors. This is the so called

active management

.

The starting point is not the selection of stocks on the basis of past data of risk, return and correlation, but the selection of assets to buy and sell

on the basis of strategies that have proven to be effective on past market data

.

It’s a scientific approach because everything in the strategy comes out from a deep study on historical data and the thorough research of the best parameters and indicators able to improve the effectiveness of the trading signals. It is all a matter of odds, of course: we buy when the odds to gain are higher than average, we sell when the odds of a pice drop are higher than average.

The expectation is that with a proper strategy it could be possible to improve the timing of trades

, so to achieve a global return higher then that of the simple buy & hold.Slide18

18

The phases of an investment

For people not used to trading, to

invest money on their own

in the 21

st

century might seem

a

very simple operation: you turn on your pc, log-in to your account, choose a stock (or another kind of instrument) and buy it.

Actually this apparently simple process is the result of a long series of steps.

How do we get an access to the market? Do we need special authorizations?

How do we decide what to buy and sell? How can we actually buy and sell?

Once that we have bought a stock what should we do? When will we sell it? Why?

These are only some of the many questions that rational investors should ask themselves

.

The investing process, indeed, is much more complicated than it might seem.Slide19

19

There are eight steps to take in any investment:

get an access to the markets

define the timeline of the investment

choose what to buy and why

decide the capital allocation for that trade (money management)

decide what to do if things go wrong (risk management)

decide what to do if things go well (profit management)

choose the proper instrument to trade

send the order to the market and monitor its execution

Trading is not that simple after all…! In reverse, it demands for a highly scientific approach

The goal is to take investments fully planned under any point of view

before

taking them

.

Everything is religiously planned in order to always know in advance how to deal with any possible outcome.

The phases of an investmentSlide20

20

Steps from 3 to 6 represent the four pillars of an investment and they all concur in the same way to success or failure in this field.

The order of apperance is not important, then: they could be inverted or mixed, and the final result would not change anyway.

Nevertheless, moreover for unexperienced investors, step number five is the most important: risk management.

Only a few investors learn how to deal with it before it’s too late.

Overconfidence

is very common in this field and always leads to financial death.

The first lecture is that

risk is the only thing that matters, since it’s the only one we can control.

One preliminary noteSlide21

21

Step 1: get an access to the markets

To trade on the stock exchange there is no need of specific qualifications, or means reserved to some category of investors; but in any case it is necessary to make a series of choices and take a series of specific steps.

The first step is obviously to open an account.

At the same time we need to be authorized to operate on the market, filling in some specific forms.

Some markets – such as the derivatives one – require the filling in of some additional forms.

In the several forms to be filled in at the subscription of a trading account there are also questionnaires about the knowledge of the clients about financial markets and products, their expectations about returns, inclination to risk, and so on.Slide22

22

The access to the markets via phone and pc dates back only a few years.

Before 1996, in fact, the access to negotiation was reserved to specialized intermediaries: to invest in the markets, private investors had to go to a bank, in person.

It was possible to make orders via phone, but within the same day the client was required to go to the bank and sign in the proper forms.

Not

all the

banks allowed clients to transmit orders via phone, though, given the risk that the client could change his mind before signing in the forms (it often happened when between the order via phone and the end of the day the price dropped…).

Some more progressive bank started recording phone calls on tape, and the law adapted in order to regulate the means of acquisiton and preservation of those phone calls.

In 1996 everything changed.

Step 1: get an access to the marketsSlide23

23

Step 1: the trading on line revolution

The easy access via pc changed the rules of the game, giving input to the fast growth of products and services.

Here the main changes:

direct access to a lot of markets worldwide for anyone

huge expansion in the supply of services (professional advice, market data vending, trading platforms, instruments for data analysis, and the like)

strong reduction in costs of negotiation

enormous growth in products (number) and markets (volumes, values)

quick development of new investing techniques

birth of high frequency trading, quant trading, systematic trading Slide24

24

Here the choice is between two different paths, that sometimes can be complementary, sometimes completely opposite: shall we invest on the long term or the short term?

It’s a paramount distinction, because the most part of the whole investing process is strictily related to the choice we take in that field,

the field of time

.

Indeed, depending on the path chosen we have then to act coherently to it, applying befitting techniques. Let’s see why.

Basicly the choice is between being

investors

or

speculators

.

The firsts look forward: they make investments aiming to a growth in capital in months and years, thanks to a growth in prices in time, defending the capital from inflation.

Tipically they invest in bonds, mutual funds, ETFs, stocks of companies they think they can rely on, because they are known to provide good yearly returns in terms of dividends and some sort of stability in prices over time, even in bad general market conditions.

How do they choose these companies? We will see it forward.

Step 2: choosing the approach to the marketSlide25

25

Speculators, on the other hand, are people who invest on a short or very short term: a few days (position traders), a few hours (intraday traders), or minutes, or even seconds (scalpers).

Now, one of the unchanging laws on the markets is that

profits take time

.

Indeed it happens quite rarely to see huge price movements in short or very short terms; this reduces critically the odds to make significant profits in each trade in a short while.

To gain we have to let the market move in the favourable direction and enough to produce a gain, and this might take some time.

Let’s see some numbers to prove that idea. Between January the 1

st

2000 and December the 31

st

2013 the stock ENI (listed on the italian stock market) made a 0.03% average daily price variation (close vs previous close), and 70% of the time the price moved within a daily range between -1.41% and +1.47%.

This means that, investing money on ENI, seven times out of ten we cannot expect to gain more than a 1.47% per day, and we need to be prepared to the risk to lose up to a 1.41%.

Step 2: choosing the approach to the marketSlide26

26

Conclusion:

it is not easy to make huge profits in a short while

, because the market doesn’t help.

Investing 10,000

euros

on ENI shares today, if things go well we could have 147

euros

(gross, because we still have to cut commissions and taxes) more tomorrow, but we could also have 141 less (plus commission and taxes).

The most likely event is to have 3 more

euros,

the 0.03% average daily profit (still gross).

What if we look one more day forward?

The

daily average price movement grows to 0.06%, and 70% of the time prices move within a range between 2.09% and -2.00%: we can expect a 2.09% profit, but we have to be prepared to a 2.00% loss.

A second unchangeable market rule

:

The

longer you stay on the market, the higher are potential profits, but losses as well

It takes time to make money, but the longer you wait to make money the more you expose yourself to the risk of losses. Like it happens in many things in life, it is all just a matter of equilibrium: we need to find the best compromise between expected profit, time sufficient to reach it, and risk connected to it.

Step 2: choosing the approach to the marketSlide27

27

As additional proof, in passing from a two days to a five days horizon the average achievable profit grows to 0.14%; 70% of time we can gain up to 3.24%, risking to lose up to 3.09%.

Nothing changes, then: the longer we stay in a trade, the higher are both the potential profit and the potential loss.

Moreover, under a historical point of view the maximum 5-day return on the stock ENI has been around 27%, while the maximum risk has been around 25%.

Now the

most important question: on

a 5-day time range can you tolerate the risk to lose 25% of your money? It’s a very low probability event, sure (it happened only once in fourteen years!), but what if it happens to you?

On a 5-day time range the probability to lose a 10% or more of your capital on the stock ENI is 0.9%, while the probability to lose a 5% or more is 6.5%. Investing 10,000 euro on ENI, not managing the risk in any way, in the next 5 days there’s a 6.5% probability to lose 500 euro or more.

Step 2: choosing the approach to the marketSlide28

28

All of this is just to prove that investments have to be managed, because things can go well, but they can also go bad, very bad. A long term (talking about years) investment not managed can lead to devastating

effects.

European and American stock markets are full of very good examples of these issues.

A third rule then:

we need to know how to manage situations

And this can be learnt only with knowledge and experience on the field.

Atop of this we need a good stock-picking methodology: if, on one hand, the longer we stay in a trade the higher the risk, it is also true the if we do not stay enough in a trade it will be very hard to reach a profit. It is again a matter of equilibrium.

How to get out of this situation? We need two things; first, define a strategy that makes us get into trades given of a high probability to be profitable. What we need is

a way to identify in advance most of those situations where the odds for a tail event are higher than average: events able to bring high profits in short time periods.

Notice that we are talking of probabilities: our trading strategy has to make us win

most of the times, not always

Step 2: choosing the approach to the marketSlide29

29

Because the perfect strategy doesn’t exist: there is no strategy that makes you always win; there is no market that doesn’t do the opposite of what everyone expects, sometimes; there is always some unpredictable event, on any market; there is no internet connection that doesn’t go down the moment you

need

it, perhaps in the moment you are going to take the best trade of the year; there is always a firm that will announce very bad news right after you clicked the button to buy it.

To win we do not need the perfect strategy, but just a good entry strategy, an effective risk management policy and a good profit management

Know what to buy and when, what to do if things go bad, and what to do if things go well.

The second element we need for success is the

leverage

.

The point is that there are only two ways to gain on the financial markets: either we invest a small money amount and catch a very profitable movement (and now we know it requires time

and it

exposes

us to

increasing risks), or we invest a huge money amount and catch a small price movement (in a much shorter time, but still with high risks and very stressfully). There are pros and cons in both choices. About the second, only a few investors can afford to invest huge money amounts on the markets, and there comes the leverage. We will see it forward.

Step 2: choosing the approach to the marketSlide30

30

As previously stated, depending on the timeline of the investment we have to choose proper strategies.

If the idea is to invest on the long term on stocks that will pay good dividends and grow in price over time we have to be able to select them among the whole list.

This result can be achieved with

fundamental analysis

.

To do fundamental analysis means to analyse the trend of the economy in general, then the sector

of our interest, and finally the balance sheet of the company of our interest, in

order to estimate the so called

fair value

(that is

the right price

) of the firm, which has to be compared to the price on the market, so

to be able

to state whether the company is under-valued, fairly valued, or over-valued.

What does all of this mean?

Basically,

it’s a matter of price Vs value

.

Prices of listed stocks move up and down continuously because of the constant action of supply and demand; these two forces are influenced by the information available on the market at some specific moment, and the expectations of operators about the future trend of prices.

It is all a matter of today’s expectations about future prices

.

Step 3: defining the stock-picking strategySlide31

31

Making it simple, who buys a stock

today does

so because he expects that the price is going to increase in time; vice-versa, who sells a stock does so likely because he expects the price is going to decrease in time.

And only

one of them will be

right!

But who of the two will be right will be known only at the end of the period, since everything happens in the field of probability.

Information, hypothesis, expectations,

probability

are

then the drivers for price movements: who buys ENI shares today does so because he has information that lead to hypothesis and then to expectations of growth in its price within some time range; moreover he buys ENI shares because the probability of a growth in price is higher than the probability of a drop (at least this is what he thinks).

How can investors form these expectations about growth? Some of them

likely use fundamental

analysis, that in the field of the stock markets works more or less this way:

1. In the first place we have to analyze the general economic and political field, in order to understand whether the system the firm operates in could act in favour of its growth or not; in this field we have to give a look to the policies in terms of taxes, labour, and the like;

Step 3: defining the stock-picking strategySlide32

32

2. Then we have to analyze the sector the company operates in: point of the lifecycle of the products or services offered by the firm, levels of competitivity, innovation spaces in terms of products and processes, investments in research and development, barriers to the entrance of new competitors, and so on;

3. Then we have to analyze the balance sheet of the company(ies) of our interest, in order to determine the key-parameters (typically returns on equity, debts, expected earnings, operative margins, and the like) for the valuation of its current and future “health state”; the balance sheet’s analysis combined to what written in points 1 and 2 has to lead to the

so called

fair value

,

that is

what should be the value of the firm

considering its actual business and the expectations about its future one; the value of the firm divided by the number of outstanding shares provides, then, the

theoretical

price per share, or fair value

;

4. finally, we can compare the theoretical price per share with the market price per share, so to inspect the possible existance of discrepancies.

If the theoretical price is higher than the market price,

then

the

market seems to be devaluing the firm

,

for some reason.

Step 3: defining the stock-picking strategySlide33

33

Example: if the fundamental analysis leads us to a theoretical price of 25€ per share, and today the market price is 19€ per share, then

it seems

that the market is devaluing the

stock,

in respect to what

should be

its right price.

But pay attention, because

there might be several reasons for that

. Anyway, sometimes it happens that the market makes mistakes in the valuation of a firm, so, if we have reasonable certainty that our analysis is correct and that the market is devaluing the company with no reason, then it could be a good deal to buy

shares of that company,

under the idea to keep them in our portfolio until they reach their fair value.

Vice-versa, if the analysis states that the theoretical price of

the firm should

be 15€ per share, then we can say that

the market might be over-estimating the firm,

for some reason. Again, if we have reasonable certainty that our analysis is correct and that the market is over-estimating the firm, then we could state that it is not time to buy

those shares

, and, if we have shares in our portfolio, it could be the right time to sell them.

Step 3: defining the stock-picking strategySlide34

34

What

could go

wrong with fundamental analysis?

Fundamental analysis is based on a perfectly logic premise: if the fair value of a company is 10€ per share, sooner or later its market price will be 10€ per share.

There are limits, though, in this approach. First of all it is not that simple to calculate the fair value of a company, and it often happens that it is not unique, since it depens on the point of view (sometimes interests too) of the analyst.

A first serious problem is given by the mass of data to be analyzed and interpreted: a “normal” balance sheet can be a volume of hundreds of pages. Multiply that mass of data times the number of listed companies – 350 only on the italian market – and it takes a lifetime to read them all. The only way to do that is to pick up a basket of stocks to be analyzed, excluding a priori on the basis of some reasoning all the other stocks.

For instance, we could decide to focus on a specific sector, or on high capitalization companies (if we are seeking for stability of returns and lower risks), or maybe the opposite, only low capitalization companies (if we are seeking for higher potential returns and can stand a higher average risk).

Doing this we can significantly reduce the mass of data to analyze, even if the arbitrary selection of stocks to analyze could cause a loss of some opportunities on the firms excluded.

Step 3: limits of fundamental analysisSlide35

35

In the field of balance sheets analysis there is also the problem of the

delay

in data; it often happens, in fact, that when new data are released they refer to several months before, hence they are not a picture of the current state of the company, but of the previous one. This can lead to a not updated picture of the firm.

Moreover, fundamental analysis is a “slow” technique, meaning that it produces returns in months or years; it often happens that as time passes new data come into play, and new events happen, events that could not be predicted in the first place; these new elements bring with them the need to review the whole analysis, since the picture is changed. It happens, then, that a correct fundamental analysis needs to be in part or fully revised before the price reaches its initial target.

One of the main issues in fundamental analysis is the value of

intangible assets

: human capital (think of an IT company: it could have the new Bill Gates within its employees and not know it), new licenses that could be registered, new productive processes that could be invented, new technologies, new products and services that could be born, and the like. How can we, for instance, estimate the expected value of a potential new product? Think of a pharmaceutical company developing a new drug to cure cancer: how do you quantify it? You first need to determine the probability for the company to actually develop that drug, when, and its potential business. And how to evaluate the risk that a competitor comes out with a revolutionary product within the next months or years?

Step 3: limits of fundamental analysisSlide36

36

The point is that in the process

we also have to account for expectations and probability of

events

.

Indeed, the

fair value of a company is not only the result of a mere accounting for numbers in the balance

sheet

,

since

it can also account for the expected future value of projects already started or ready to be started; in othe words, it happens that the current market value of a stock already

discounts expectations

of future growth or downturn.

A stock quoting a price lower than its fair value could be the expression of fears of investors for expected future hard times for the company. Similarly, if a stock quotes a price higher than its fair value we could state that the market is discounting in advance a future growth of the firm.

In a few words,

the market price accounts for the past, the present and the future (known, realiable, or

even just

probable)

.

In other words,

prices discount (almost) everything

.

Step 3: expectations and probabilitySlide37

37

Expectations are usually a function of personal opinions of analysts, both in terms of the effects (on prices, of course) of specific events and of the quantification of those effects.

The value of a new model for an automotive company, for instance, could be hard to quantify and the valuation of expected earnings could be significantly different among analysts.

The birth of a new technology could represent a competitive advantage only for young and dynamic firms, and mean earnings downturns for

firms

that find hard to modernize themselves; the quantification of these issues could not be the same among different

analysts, because of their personal opinions.

But looking at these issues from another point of view we realize that if it wasn’t so, if there were no asymmetries in the valuation of economic matters, there would not even exist the markets!

If, for instance, all the analysts agreed that the fair value of a stock was 19€ per share, who would ever sell it for less than 19? And who would ever buy it for more? Nobody, of course! The market would not exist, and no trades would take place.

Different expectations

and disagreements are the natural fuel

for the asymmetries in information that make possible trades on the exchanges: trades made by sellers who get rid of their shares for their reasons, and buyers with opposite expectations.

Step 3: expectations and probabilitySlide38

38

These asymmetries are originated basicly by two factors:

different levels

of

access to

information

, in terms of time (not all investors receive the same information in the same moment),

and capability to properly analyze and interpret that information

.

Give the same balance sheet to ten different analysts and you will very likely have ten different valuations of the fair value of the company.

Asymmetries in the availability of information often cause masses of investors

to take wrong decisions

, buying shares that already grew too much, and selling them when prices dropped significantly, and a bounce is at stake.

Generally speaking,

the masses do the opposite of what they should do

.

Americans have many proverbs that can describe the behaviour – sometimes apparently irrational – of the markets. One of the most famous is

“buy on bad news, sell on good news”

. In other words, do the opposite of what it might seem reasonable to do.

Where does the irrationality of the markets come out from?

Step 3: expectations and probabilitySlide39

39

Step 3: about the irrationality of the markets…Slide40

40

When companies announce explosive earnings, it sometimes happens that their prices drop dramatically, out of any common sense.

Investors usually buy shares in that moments, according to what seems to be logic: good news, then good perspectives, then we buy.

But that happens usually because of misleading advertising,

news in red letters on papers

When the “smart money” (banks, mutual funds, insurance companies, and the like) needs to sell huge amounts of shares, it needs someone willing to buy them… The red letters in papers are made to attract investors, so that the “sharks” can sell.

Why? Because they already took advantage from the news, either because they knew it long before or because they foresaw something before anyone else. Going deeper into the matter, it often comes out that the price increased significantly weeks before the announcement of explosive earnings.

Who sells today bought long time ago, when prices were low because no investors were attracted by that company. Now that everybody knows it’s time to sell, taking advantage of the river of buyers generated by the news on papers.

To gain we need to be one step ahead all the others

Step 3: the trading edgeSlide41

41

Maybe a little step, but always ahead. That’s the only way to win; when news are public and can be interpreted in one way only, they are worthless, since they don’t carry any advantage.

Once again, what allows to make profits

is the asymmetry in information

.

The advantageous asymmetry can be achieved basicly in two ways: the first is to gain early access to the information; this is the hardest way, and it is also a phelony, because it’s insider trading, forbidden by the laws, national and international (information should be available for all the operators at the same time, in a perfect world…).

The second way is to be able to take advantage of new data before the others, seeing (and foreseeing) things that nobody else, or only a few people, can see. But to do that we need skills and tools better than average.

A paramount element is the testing of the effectiveness of our trading techniques over past data:

we cannot invest money on the basis of trading rules that haven’t been tested on historical data and proven to be effective

, givin’ us a sure competitive advantage; read it

systematic profits

.

We need, then, to translate our investing techniques into sets of rules clear and well-defined, that can be applied mechanically as specific sets of conditions show up.

Step 3: the trading edgeSlide42

42

An example? A coded strategy based on fundamental analysis could be “buy a stock when the board of directors announces the payment of a dividend higher than the previous; sell one week before the payment of the dividend”.

Does it work? The only way to answer that is to take data of all the listed stocks over the past 10 years or more and check the results that we could have achieved. The longer the time period under analysis, the higher the number of observation, giving the analysis a much higher

statistical significance

(read it

reliability

, a measure of the probability that what happened in the past will keep happening in the future).

We have to do this,

because as stated before we have to define a trading strategy that gains most of the time (always it’s not possible); how can we know that our strategy gains most of the time if we do not test it?

To test the effectiveness of a trading strategy is something very popular these days in the trading field. You can hear people talking about quant trading, algorythmic trading, quantitative, systematic, automatic. All similar terms, even if not synonims.

Step 3: systematic tradingSlide43

43

The common denominator to all of them is that a systematic trading strategy – which means doing always the same things any time a specific set of conditions shows up – can be managed in two ways.

The first is

manually

: in the moment a trading signal shows up we manually send an order to the market; the second is

automatically

: we program a computer so that when a specific set of conditions shows up it buys or sells something on its own.

The importance of the coding of trading strategies is a concept that sooner or later becomes perfectly clear to any trader,

because

investing

money without knowing whether our approach is winning or losing in the long term leads us to an unbearable stress

.

It usually takes just some negative trades in a row to make investors abandon any improvised strategy. It is mostly a problem of emotions, ranging continuously from euphoria – when things go well – to fear – when they go wrong, passing through a wide set of intermediate states

.

The only way to keep emotions on a lead is to code trading rules so to always act in

cold blood

. Rules tested on past data in order to define trading strategies able to produce consistent gains always keeping risk under full control. Operative rules can be based on fundamental analysis or technical analysis (which we’ll see now), or even a combination of both.

Step 3: quant tradingSlide44

44

Technical analysis was born at the end of the XIX century thanks to the studies of Charles H. Dow, an american, founder of “The Wall Street Journal”; since then it developed – and keeps developing today – thanks to the contributions of many researchers.

The primary goal of technical analysis is analyze financial markets and recap all the available information in a few indicators – the price most of all – that can be represented on a chart. The idea is to take operative decisions based only on the chart, without any concern about balance sheets, the general state of the economy, policies about labour, taxes, import and export, and the like.

Technical analysis is based on

three primary principles

:

1. all the information available in the markets at some time is

discounted in

the current price;

2. history repeats itself: under the same conditions, what happened in the past will happen again in the future, with similar effects;

3. prices do not move randomly; on the opposite, they draw trends that can be used to make money.

In other words,

prices move in patterns that tend to repeat in time

; we can take advantage of these patterns to open positions under the expectations that trend will behave coherently to what happended in the past under the same initial conditions.

Step 3: technical analysisSlide45

45

Example

Suppose to observe that any time the price makes a new annual maximum it keeps moving up the next year: we can say we found a repetitive pattern in the market, that can be used to gain money the next time it shows up.

We are always in the field of odds, of course: technical analysis is based on the observation of patterns that tend to repeat on a statistical basis: what happens today

should

make the price move in some specific way, but nobody can give us full guarantee that it will happen.

Any strategy based on technical analysis requires then testing on past data: we need to code the operative rules of our strategy and apply them on historical data. If, after a backtesting, the strategy gives proof of efficiency, restituting a satisfying profit over time, then it could be reasonable to use it on future data, assuming that what has proven to be effective on past data will be effective on future data too.

We will see some backtesting of trading strategies forward.

Step 3: technical analysisSlide46

46

46

Another very famous american proverb is “

trend is your friend

”. Its meaning is quite simple: if the price of a good is increasing, this means that the market – on average – thinks that the real value of that good is higher than its current price.

The reason underneath that behaviour is not important: what matters is that the market is the sum of all the expectations of the operators active in it; if positive expectations are stronger than negative ones, the only reasonable conduct is to follow the current, that is adapt to the positive view of the market.

It’s time to buy.

This reading of the market remains valid until proven otherwise: when negative expectations become stronger, it’s time to sell.

There is a clear, unavoidable, limit in technical analysis, then:

it never

anticipates

the future

; it just follows the current, adapting to it. Indeed, it is never possible to state that the price is going up or down without

watching

it going up or down.

It is impossible to anticipate changes in trend directions: we can only adapt to the clear trend in the price. It is not possible to buy on a minimum or sell on a maximum, then, but by chance.

The basics of technical analysis: the trendSlide47

47

47

Technical analysis can only tell to buy a stock that has already risen from a relative minimum, and sell a stock only with some delay, after it has already dropped from a relative maximum.

Timing is crucial

, then: the more accurate our investing strategy, the eralier it could identify the birth of a new trend in the price, and take advantage from it.

As we’ll see, it is possible to improve investment strategies using several types of indicators; some of them can be used to anticipate possible future changes in the price direction, allowing investors to

reduce

(not eliminate)

the operative delay

.

Everything is based on the price charts: they are the key element to start from.

Charts are just price representations over some time period; the two variables on the axis are then

price and time

.

Time can be in days (that’s the most common way, called

end of day

), but also in weeks, months, quarters, or years; or even in fractions of days: 240 minutes, 120, 60, 30,15, 10, and so on down to a minute (

intraday charts

).

On any market stocks and other instruments are quoted in time bands and according to specific rules set by the Exchanges; on the italian stock market, for instance, stocks are quoted daily, Monday to Friday, 9am to

5.30pm

.

The basics of technical analysis: the trendSlide48

48

48

This is the so called

continuous phase

: there are no stops to negotiations in this phase, unless particular events happen.

Befor the continuous phase there is an opening auction (8 to 9 am); in it buyers and sellers show up on the market with their buy and sell proposals. Orders are collected by IT systems that match them according to specific rules in order to reach some sort of equilibrium price, that is taken as opening price (we will see these dynamics forward).

Then trades happen all day long, generating waves that will lead to an intraday maximum and an intraday minimum (not necessarily in this order).

At the end of the day trades stop and a final price is recorded: the price of the last trade of the day. That price can be seen as some sort of equilibrium point for that day: it is the result of all the known facts and the expectations for that stock in that moment.

Indeed at the end of the continuous phase a closing auction takes place. It lasts only a few minutes, but leads to a closing price that

fulfills

some special rules.

To be thorough, there are two more possible closing prices: the official and the reference one. Without going into too much details, they are averages of the prices, weighted on the traded volumes, used to give more information to specialists. The reference price is also used when the closing auction does not provide a closing price (it happens, sometimes).

The basics of technical analysis: the trendSlide49

49

49

Summing up, for any day there are

four key-prices

pointing out the dynamics of supply and demand: open, high, low, close.

Given these four values we can build several price charts; each of them carries a different informational power and can be read in several ways.

Here we will speak of the line chart, the bar chart and the candlestick chart. There are other types too, but they are out of our field of interest for this course.

If we build a chart using only closing prices for any time period (usually a day), and then we link all the closing prices with a line, we get a so called

line chart

(see next slide).

The basics of technical analysis: the trendSlide50

50

50

The line chartSlide51

51

51

The chart on previous slide shows the FTSE Mib Index, including the 40 main stocks

(the so called

blue chips

)

listed on the italian Stock Exchange, between the end of May and the beginning of September 2014.

How can we read this chart? What information can we get from it? How can we use this information to gain money?

In order to answer these central questions we

just have

to

understand

the essence of technical analysis, in the first place. Let’s recall three major concepts stated before:

1. if the price of an asset grows, it’s because there are more optimists than pessimists,

more buyers than sellers

;

2. to profit from price movements we need to

identify the trend and follow it

;

3. we have to be prepared to deal with

some degree of operative delay

, since we cannot tell that a market is growing or dropping if we do not see it growing or dropping for a while.

A very important rule, then:

follow the market, and don’t try to anticipate it.

The line chartSlide52

52

52

As we identify the trend, let’s say up, we have to buy and follow it, then wait to see what happens. In the moment the trend changes direction we have to understand it and adapt our behaviour, selling to get out of the trade.

Now we have

two

issues to deal with:

how to identify the trend

, so to be able to follow it, and

how to spot the end of

that

trend

, so to be able to get out as prices start to drop. As previously stated, timing is crucial: the sooner we are able to spot the beginning of a new up trend, the lower will be the price we pay; the sooner we will be able to spot the end of the up trend, the higher will be the price we will sell for.

Timing is all

, as Americans say!

Let’s look back to the chart and let’s try to spot the trend. Where is the price going? That might seem a trivial question, but to answer that in an unquestionable way we need to define a clear, scientific, rule.

Indeed it is not possible to give an answer to that question without introducing a second variable in the model:

time

.

This because the direction of the trend is not always the same: prices move in waves, that in each time window – on average – move upward or downward; then, at some point, they change direction, and then prices move to the other side. And so on.

Price and timeSlide53

53

53

A way to answer the question about the direction of the price is to compare the final price with the initial one: nowhere! Indeed the final price is more or less equal to the initial one. This is said to be a

sideway trend

: no direction. Another way could be to divide the chart in zones, like this:

Price and timeSlide54

54

54

Basicly, price waves generate maximums and minimums that can be higher, equal or lower than the previous ones; this information can help us define the general trend of the price.

Indeed we can synthetize these ideas in the first postulate of technical analysis:

If two consecutive relative minimums and maximums are growing, the trend is up; if they are all more or less at the same level, the trend is lateral, or non-directional; if they are decreasing, the trend is down.

In zone 1 the trend is up, down in zone 2, up again in zone 3.

Now we know how to spot the trend.

Later on we will also learn how to take advantage from it, and how to identify changes in the price direction.

Price and timeSlide55

55

55

Looking at a line chart it seems quite easy to spot minimums and maximums: a minimum is a point set between higher points on both sides; a maximum is a point set between lower points on both sides:

Minimums, maximums and trendlinesSlide56

56

56

They are all relative minimums and maximums; they are also called

short term

minimums and maximums. Not all those points are significant, by the way: looking closer, some of them seem to be more important than others, since they are easier to spot:

Minimums, maximums and trendlinesSlide57

57

57

Under another point of view, we can state that those points are more important because they mark significant tendencies.

Linking minimums and maximums with straight lines we can finally obtain tendency lines, or

trendlines

.

Let’s focus on the chart on previous slide and remember the first postulate of technical analysis. In the chart we have three key-points: two minimums and a maximum. The two minimums are increasing (the one on the right is higher than the one on the left), hence

we can state we are in an up trend

.

Linking the two minimums with a line (see next slide) we get our first trendline.

A trendline linking two growing minimums defines an up trend, then. What is the utility of the trendline?

It delimits the direction of the price, and can tell whether the price is still moving in the same direction or the wind is changing.

Minimums, maximums and trendlinesSlide58

58

58

In technical terms, the line linking growing minimums is called

support

. Indeed it has the

attitude to

support the price, keeping it oriented to the upside.

Minimums, maximums and trendlinesSlide59

59

59

Now we have spotted the trendline we need to understand

how we could use it to take right investing decisions

.

The idea is that if the support can keep the price moving to the upside, pushing it up each time it is tested in the future, then it could tell us when to buy.

What we have to do, then, is wait for the next test of the trendline and buy in the moment of touch.

See next slide: the points A and B are the two minimums seen before. In the point C the price leans on the support and bounces to the upside. Buying in C has proven to be a good choice.

Will it happen again in the future? We can only wait and see! And it happens again in D.

Minimums, maximums and trendlinesSlide60

60

60

It seems that the idea could be valid: each time the price touches the support it is pushed back up; the trend is still upward, since all the new relative minimums are higher than the previous ones and after each new minimum the price moves up seeking for higher maximums.

Minimums, maximums and trendlinesSlide61

61

61

There are some problems in this approach, though.

Two, above all: what to do if the support doesn’t work (the price cuts it to the downside), and how to manage open trades.

The first issue concerns

risk

management

:

what to do when things go wrong; we buy on the support and the support doens’t work anymore.

The second issue concerns

profit management

: what to do when things go well.

Indeed, any trade is made of three fundamental elements: choose what to buy and when, set the capital per trade, decide when to close the trade, either it gains or loses money.

Sooner or later we will have to get out from a trade! We will deal with all these matters forward.

Minimums, maximums and trendlinesSlide62

62

62

Now let’s see what happens when a support doesn’t work anymore: on a new test the price cuts it to the downside instead of bouncing back up:

Weak supportsSlide63

63

63

In the point E the

dream is broken:

the price tests again the support and cuts

it to the downside.

If we buy in E assuming that the price will bounce back to the upside we will have to deal with our first big problem: what

should we do now

?

Let’s move on to resistances before answering that question.

As

it is

possible to link growing minimums with trendlines and draw supports, it is also possible to link decreasing maximums and draw other trendlines, called resistances.

A

resistance

is a trendline that keeps the price oriented to the downside: each time the price touches the line from below it is pushed back down.

See next slide for an example: as we link the two points A and B we have the resistance; it pushes back down the price four times then, in C, D, E and F.

ResistanceSlide64

64

64

Remember

what was stated about trends: when you spot a trend you have to go with it, following it until it changes direction.

ResistanceSlide65

65

65

In the moment the support is broken we have to deal with two issues: how to manage the risk of losses and how to read the chart from now on.

We will take care of risk management forward; let’s first discuss the second issue.

Empirical observation of the market behaviour leads us to state that:

a broken support tends to become a resistance; a broken resistance tends to become a support.

This is known as

the state change postulate

: when a trendline is broken it usually changes its state; what once pushed up the price

now

pushes it down, and vice-versa.

Look next slide: after the breaking of the support in the point E the price drops, then bounces and tests again the ex-support from below, point F (and again soon afterward), and this time the support pushes the price on the downside: it became a resistance!

ResistanceSlide66

66

66

We can then unquestionably state that the uptrend is over.

ResistanceSlide67

67

67

We can then sum up all these concepts in a

new

postulate:

an uptrend (a downtrend) can be considered over when the support (resistance) is broken and a next test confirms its state change.

That’s what happened in E and F on previous slide: breaking of the support, price drop, bounce, new test, state change confirmation.

There is not much we can do now: there are no signs of a possible new uptrend, so we have no buy signals. All we have to do is wait for new trading

opportunities

.

We will then wait for a new support to be created, and for new tests afterward. And so on.

The end of a trendSlide68

68

68

The strategy to buy on a support in an uptrend can be classified as the strategy

to buy a strong stock in a moment of temporary weakness

.

Recalling all the ideas seen so far, in fact, we can state that after spotting the two growing minimums sequence and the maximum between them, the wait for the new test of the trendline in the point C means to wait for a moment of weakness within the ongoing up trend.

This temporary weakness can be exploited to buy according to the current up

trend, assuming that the price will move up again soon.

As previously stated, prices always move in waves; the general trend depends on which waves are wider.

Look at the chart on next slide: the waves go alternatively up and down, but the width of upwards –

impulsive waves

– is higher than that of the downwards –

corrective

waves

– then the trend is up.

Buy the correctionsSlide69

69

69

Impulsive and corrective waves

impulse

correction

impulse

correctionSlide70

70

70

To buy on weakness is coherent to the logics of technical analysis, under some point of view: it assumes that the trend will keep going up until proven otherwise, because that is the information the chart is giving us.

Under another point of view it is incoherent, though: since technical analysis cannot anticipate the future, who can tell us in advance that the support will keep pushing up the price on a new test? No one, of course.

Indeed, assuming today we are in the point C on slide

60,

nobody can tell us that the price will bounce for sure tomorrow: nothing is certain and nothing can prevent the price to behave in C as it will behave in E later on.

The point is that it is not correct to buy in C and D, but

the day after those points

, when we will know for sure that the price bounced on the support and it will likely keep bouncing

.

We have to allow some

delay

, then, but this prevents us to lose in the point E: in the moment the price cuts the support instead of bouncing, we do not buy.

Buy the correctionsSlide71

71

71

And by the way we can also take into consideration another way to do the job.

Till the price keeps moving up, the support plays a very key role: if the price keeps moving above it, bouncing on any new test, the trend will stay up.

Now let’s reverse this statement: given a resistance, what we expect is that it will keep the price going down until it will be broken to the upside. If, on one

hand,

the breaking down of a support states the change of the trend from upward to downward, on the other hand the breaking up of a resistance

states the change of the trend from downward to upward

.

This leaves room for a radically different trading strategy:

to buy on strenght

, not on weakness.

Look next slide.

Another way to do thatSlide72

72

72

After the breakdown in E and the test of the support in F, the price draws two new decreasing maximums, G and H. The resistance linking them is the new guideline for the trend, that will be downward until a breakout shows up.

Buy on strenghtSlide73

73

73

And later on we have a breakout of the resistance trendline. This is a symptom of a possible trend change, from down to upward. Where this breakout will lead us, we can’t tell in advance; in this particular case it leads to a powerful growth in the price:

Buy on strenghtSlide74

74

74

One very important feature in trend analysis refers to the expected duration of the trend: it’s a paramount element, since the longer the lifetime of the trend, the higher will be the profit achievable following it.

Emipirical observation of the markets leads to a new postulate of technical analysis:

On a statistical basis, the duration

of a trend is

directly

proportional to its width and inversely proportional to its slope.

The wider the waves and lower the angle of the climb, the longer will likely last the trend; vice-versa, the steeper the angle of the climb and the narrower the price waves, the shorter will likely be the trend.

In the chart on next slide you can see the FTSE Mib index on a quite long time window, about two years. In region 1 you see a strong uptrend: the price increases significantly in a short while and the few corrective waves are very short; but this trend lasts only two months. Another very strong trend in region 2: it lasts only three months, with only a few and very short corrections. In region 3, instead, we see a very long downtrend, about 15 months, with a very low average slope and wide corrective waves (to the upside this time).

Expexctations about the lifetime of the trendSlide75

75

75

Expexctations about the lifetime of the trendSlide76

76

76

The approach presented in the past slides allows investors to take any chart and read it perfectly,

but

in

retrospect

. Indeed,

in applying it in real time it

raises several problems, that make it much more complicated than it might

seem:

1. which trendline to draw on a chart full of so many minimums and maximums is not so trivial;

2. depending on the time window of the chart, the interpretation can change significantly; this leads to the drawing of

many

different trendlines depending on the duration of the trend under analysis;

3. two different

analysts

in front of the same chart could likely draw different trendlines, since the importance of specific minimums and maximums is often subjective;

4. if, in order to draw trendlines

with

a high degree of importance, we seek for lines that get tested precisely a high number of times (assuming each new test gives them more importance), we can very soon realize it will be a very hard job: we can link many pairs of minimums and maximums, but to be able to align three or more minimums or maximums is much harder

About point 4, all the points in the charts on previous slides are not as precise as they seem: it is evident looking closer.

Limits of the trendline based approachSlide77

77

77

One of the paramount argumentations in the field of trends and trendlines based trading is whether the price behaviour approaching to trendlines and after the breakouts is random or not. Under some point of view, even on the chart of the body temperature of any human being it could be possible to draw trendlines. Could we state, on such a chart, that the breakout of a resistance would likely lead to a high temperature? Sure not!

What makes different, then, the chart of a body temperature from that of Cocoa, Telecom Italy, Gold, Oil, the Nasdaq Index or the german Bund?

The answer lies in the psychologic dynamics acting on prices: body temperatures move as a response to physical matters, out of our control; market prices of financial assets and traded goods move in response to human behaviours, which in turn are moved by news, sure facts (that can be interpreted in different ways), deceits, expectations, emotions.

When, for instance, on a chart we can draw a resistance line that can be seen by masses of investors set all around the world, we can expect that the breakout of that line will trigger a powerful price movement, appealing rivers of buyers, pushed to buy under the expectation that the breakout would be just the prelude to new highs soon to come.

Emotions play a central role in technical analysis.

Do prices follow a random walk?Slide78

78

78

Moreover, deep studies made by famous traders such as Larry Williams pointed out that

prices of financial assets do have memory

: what happens today is the result of today’s news and yesterday’s trend; in other words, what happened yesterday affects what happens today.

And as a consequence, what happens today affects what will happen tomorrow.

Indeed, Williams gave proof that the odds to have a positive (negative) day today are higher than average if yesterday was a positive (negative) day too. This happens because when the market has a positive or negative view it lasts more than one day, most of the times. If prices, day after day, were independent, the odds to have a positive day after a positive day should be 50-50 (just like the flipping coin game)!

Williams also gave proof that to buy a stock on any day of the week is not the same: some days are most likely positive than others… The same happens, in different days, for bonds!

Some months of the years tend to be statistically more positive or negative than others (see the January effect, “sell in may and go away”); years that end with 5 are statistically very positive for stock indexes. The list of these curious “random” phenomena is very long…And the conclusion is only one:

patterns do exist

.

Do prices follow a random walk?Slide79

79

79

Only a couple more things before moving on:

1. it sometimes happens the trend is non-directional, or

lateral

; this happens when pairs of minimums and maximums are more or less at the same hight; they are called

double minimums and maximums

. In such situations trendlines are horizontal and both the support and the resistance have the same importance: the first that is broken will likely point out the next

trend;

2. breakouts of trendlines not alway lead to trend reversals; it often happens, indeed, to testify so called

false breakouts

: the price breaks a line and soon afterward it comes back in, denying the idea of a new directional trend opposite to the previous one. It is part of the game, though: as previously stated, there is no strategy always right; what matters is that a strategy is valid on a statistical basis…

Final considerationsSlide80

80

80

In

studying the trend in different time windows it sometimes happens to be puzzled about the correct reading of the price

direction.

Indeed

, in any trend there happen to be corrective waves; and as a matter of facts, every wave is just a correction of a previous wave heading to the opposite direction.

This is just to reiterate that analysis must always be put in the right time

context.

A

short term uptrend, for instance, can just be a bounce in a mid term downtrend, or vice-versa. These dynamics can lead to some form of indecision about the direction of the trend, because we can be undecided about which wave is more important than others. To take investing decisions in similar situations can lead to beginner’s

mistakes.

To help us in the reading of charts we can use several indicators that can be plot on charts together with the price.

By the way, in the field of technical analysis the main goal of any strategy is to identify as soon as possible any change in trend, so to have the chance to get in and out with a timing better than that of other traders. A good help here comes from moving averages.

Puzzling chartsSlide81

81

81

To calculate the average of some number X of daily closing prices we sum all the X terms and we divide the result by X. If on any new day we start over and we calculate a new average, cutting off the oldest term and replacing it with the new one, we get a

moving average

.

Consider

a three days moving average, for instance: we first need three days to calculate it. As we have the third closing price we calculate the first mean. Then, the next day we remove from the data series the

closing price of day 1 and

substitute it with the closing price of day 4. On day 3, in other words, the mean is the average of days 1-2-3, while on day 4 the mean is the average of days 2-3-4. And so on. Every day the moving average moves, but it is always based on the same number of data.

Moving averages are clear

trend indicators

. Despite their simplicity, in fact, they are widely spread among professional investors; their best merit is they smooth the noise of the price waves, pointing out the only relevant information: the direction of the price. When the waves are rough, moving averages make things much more linear.

An example will make things clear. The chart on next slide shows the stock A2A (the electric company of Milan), listed on the italian market, in the period between February and June 2014. Aside the price line – the dark continuous line – there is also a moving average, the dotted line. It’s the 20-days simple moving average, or SMA20.

The number of days is the so called

pace

of the average. The 20-days SMA is very common in the field of trading, since it’s more or less the duration of a working month.

Moving averagesSlide82

82

82

The choice of the pace of the average is not random. Later on we will discuss the crucial – and very delicate – matter of the choice of the best pace for the moving average.

Moving averagesSlide83

83

83

Notice how the moving average moves in the same direction of the price, most of the time, but in a much more linear way:

it smoothes the noise

of the market keeping its direction.

Here

a crucial fact

: when the trend is up, the moving average lies below the price; when the trend is down, the moving average is above the price.

Focus on the relative maximum right on the left of the point B: the moving average 20 days calculated in that moment is the average of the closing price of that day and the previous 19, all lower than that; as a consequence, the average is lower than the current price. In the same way, if we focus on the relative minimum right on the left of the point C, we can see that the moving average 20 days is the mean of the closing price of that day and the previous 19, all higher than that: the mean is higher, then.

Summing up, in downtrend the moving average is higher than the price, in uptrend it is lower. These dynamics are perfectly clear on the chart: on the left side we have an uptrend and the average is below the price; in the central part the trend is negative and the average is above the price; on the right side the price rises again, and the average is below it.

Moving averagesSlide84

84

84

And finally

a paramount result

: when the trend changes direction, the moving average cuts the price: sometimes

from below to above,

sometimes

above

to

below.

And recalling the relationship between price and moving averages, the conclusion is that when we see a cross between the price and the moving average

a trend change is very likely going on

.

There are two possible cuts: when the price cuts the moving average

from below

to

above,

points A and C on the chart, the trend changes direction from negative to positive; it is called

golden cross: it’s time to buy!

Vice-versa, when the price cuts the average

from above

to

below,

points B and D on the chart, trend changes from positive to negative; this is a

devil cross: time to sell!

Notice how the average tells the investor to buy in A, right before a powerful uptrend, and then to sell in B, just before a deep drop; then to buy in C, where a new uptrend is at stake, and finally to sell in D, when the trend is getting weak.

Notice also how the average acts sometimes as resistance, some days before the point C, or support, just a few days before the point D.

Moving averagesSlide85

85

85

Finally, to use a moving average requires the acceptance of some degree of delay. Indeed it never tells you to buy on a minimum or to sell on a maximum: it provides buy and sell signals with a remarkable degree of realiability,

but always late

, never on key reversal price points. It’s part of the intrinsic nature of technical analysis: ther is no way to predict maximums and minimums.

So far we used the 20-days moving average. What would have happened using an 8-days or an 80-days? The chart on next slide shows again the stock A2A, and three moving averages: the 20, the 8 and the 80-days.

What changes with the pace is clearly the position of the moving average in respect with the price: the higher the pace, the higher the distance between them.

Moreover, what changes together with the pace is

the reactivity of the moving average

to price changes

.

This can lead to advantages and disadvantages.

The right pace for the averageSlide86

86

86

8-days moving average

20-days moving average

80-days moving average

The right pace for the averageSlide87

87

87

As we can see, the 8-days moving average provides more promptness to operative decisions: the first buy signal, point A using the 20-days MA, comes a few days earlier. The price paid is lower, then.

At the same time, though, the 8-days MA gives the exit signal too early in respect with the 20-days MA, since a devil cross shows up mid May; and soon afterward there is a new golden cross that makes us buy again, to sell on a new devil cross a few days later (a bit on the left of the point B). Here we have the indication of a new possible downtrend a few days earlier than if we were using the 20-days MA.

Between late April and early May, the 8-days MA produces a series of useless trades: in and out more or less at the same prices all the time. A new good trade happens late May: the 8-days MA provides a buy signal a couple of days earlier than the 20-days MA, hence for a better purchasing price; the sell signal comes around June 10, for a price more or less equal to that comes out using the 20-days MA, but ten days in advance.

The right pace for the averageSlide88

88

88

Summing up, the shorter the pace of the moving average, the prompter are the buy and sell signals, and the more convenient are purchasing and selling prices. This approach has also a drawback, though: it sometimes produces false signals, wrong trades, that cause operative costs and no gains (sometimes losses too).

Now look

at the

80-days moving average. It delays

too

much the trades, penalizing the trader. Look, for instance, at the first part of the chart: the golden cross shows a few days later than in the other cases, making the trader buy for a worse price. The devil cross – the signal to get out – then comes

too

late, telling the trader to take profit when the price has already fallen dramatically from the relative high. In the rest of the period there are poor buy and sell signals. A complete disaster.

Given the fact the 80-days MA is not a good choice, which should we take between the 8-days and the 20-days? It is hard to give an answer without a statistical test: we need a computer to

get that

answer. Indeed, nothing can tell us that the 10-days is not even better than both the 8 and the 20! And what about the 12, the 14, and so on?

The right pace for the averageSlide89

89

89

Here we see how important the backtest of a strategy can be. Only a computer can test all the possible paces for the moving average for each listed stock,

bond,

currency, stock index, commodity, mutual fund, and the like, and tell us which set of values would have provided the best result –

the maximum reward on risk ratio

– on historical data.

There are several softwares available these days for this purpose, such as Multicharts, Metastock, Tradestation, Visual Trader, or Pro Real Time; with a bit of Visual Basic it is also possible to test strategies in Excel.

On next slide we can see a study on the stock A2A between July 1998 and August 2015. The purpose of the study was to find the best pace for the moving average. In order to do that, the software Metastock has been used; the idea was to buy on a golden cross and sell on a devil cross.

Metastock has been used to test all the paces between 4 and 30 days, with a step of 2 days per test. Given an initial capital of 10k euro, and using all the available money for each trade (here we are taking a step in the field of money management, which we’ll see soon).

The right pace for the averageSlide90

90

90

For now let’s just focus on columns “% gain” and “OPT1”: the former reports the gross percent profit for each pace, the latter reports the pace for the moving average that produced that percent profit. It’s clear how the best choice on the stock A2A is the 8-days (results are sorted top to bottom). Indeed it would have produced a gross total 69% gain, while the 20-days, for instance, would have produced a 5.4%, and the 30-days a -8.2%.

The correct evaluation of trading strategies, indeed, is a very complicated matter, since the parameters to be considered are several; we will speak again of these issues later on.

The right pace for the averageSlide91

91

91

What matters now is to properly understand the meaning of what we just did: we used a personal computer to tell us on historical data and on a statistical basis what is the best way to invest our money. The result is a simulation of what could have been gained in the past trading in some specific way or another.

To trade in the future in the same way means to assume that history will repeat itself, and what worked in the past will keep

working in

the future,

similarly.

Indeed, this is one of the three principals of technical analysis (see slide

44).

Is it possible

to do

this

strategy

optimization according to some criteria that could provide the trader a good confidence that what happened in the past will likely happen again in the future with similar results?

The answer is positive, as we’ll see. We are entering the world of systematic trading.

The right pace for the averageSlide92

92

92

Previously it has been stated that a trading day starts with an opening auction and an opening price, then there is a continuous phase that leads to an intraday minimum and maximum, and finally there is a closing auction to determine a closing price.

A line chart shows the dynamics in the closing price only, losing all the other information. Indeed, the only closing price is an incomplete picture of a trading session: how can we now how the price moved during the day looking only at its last value? Was it a quiet session or a lively one? In case it was a lively day, who took the lead of it? Bulls (the buyers) or bears (the sellers)? Is the maximum far from the minimum, or are they quite close? Is the close higher or lower than the open? How far are open and close?

All these information can be very important in the trend analysis and in the decision taking process, since the inner dynamics of a trading session can provide useful tools for the interpretation of the current trend.

How can we go beyond the line chart, then? The answer comes from the bar chart (see next slide).

The bar chartSlide93

93

93

The bar chartSlide94

94

94

For each trading session we no longer have just a point, but a figure, made of a vertical bar and two small horizontal scores, one pointing to the left, the other to the right.

The horizontal score pointing left reveals the opening price, the one to the right the closing price; the two extremes of the vertical bar point out the maximum (top) and the minimum (bottom).

For each trading session we no longer have just one information, then, but seven: open, high, low, close, distance between open and close, position of the close in respect with the opening, distance between maximum and minimum.

The distance between open and close

shows how much the market’s judgment about the fair price of the asset is changed during the day: if, for instance, the opening price of the stock ABC for some day is 8€ and the closing price is 7.2€ (-10%), this means the valuation of the company ABC in the minds of investors is changed dramatically, for some reason.

The bar chartSlide95

95

95

The position of open vs close

tells if the day has been positve or negative;

the distance between minimum and maximum

, alias the

daily range

,

is a measure of volatility of the asset,

that

is a measure of

risk and opportunity

.

Volatility

of the prices of a financial asset is a key parameter for the evaluation of the price dynamics over some time frame. We will speak again of this central issue later on. Bars can be daily, weekly, monthly, quarterly or yearly; but also intraday: 4 hours, 2 hours, 60, 30, 15, 10, 5, and

even down to

1 minute.

Before discussing interpretation rules for bar charts we move on to the japanese candlestick chart, that is just a variant of the first.

It is said that the candlestick chart dates back even to the 17

th

century; in the USA and Europe, though, it became popular only in 1994 thanks to Steve Nison, a famous american trader.

The idea is to represent a trading day with a figure, on the basis of the four key-prices open, high, low and close. In order to immediately point out the direction of the day, candles can be of two different colours: white (or green) for positive sessions, black (or red) for negative sessions.

The bar chartSlide96

96

96

Instead of a bar, this time a rectangle is used, combined with two vertical lines. The former is called

body

, the latters are called

shadows

. The body is white or green when the close is higher than the open, black or red in the opposite case. Once again, seven inputs in just one sight, then.

Moreover, the specific shape of a candle, and its position within the overall short term trend bring with them much more information power, as we’ll see in the next slides.

Candles can be interpreted on their own, in pairs, in groups of three or more.

Candlestick charts

High

Close

Open

Low

High

Open

Close

Low Slide97

97

97

The width of the body and its colour

point out the tenacity of the price in a specific direction along the trading session: a wide white body, for instance, means that

demand was stronger than supply all day long,

pushing relentlessly the price up.

A

long white candle

means strenght, then, while a

long black candle

means weakness (see next slide).

Short bodied candles

reveal indecision: it seems the market doesn’t know what to do.

Here the shadows can be crucial.

Long shadows

, indeed, tell us that

demand or supply pushed the

price

up or down,

but at some point

the other side of the market took the lead.

Long shadows on both sides

unveil a fierce fight between buyers and sellers took place, ending in nothing (see slide

99).

Long bilateral shadows and short bodies

, like in the circled candles in the chart on slide

99,

mean the price is substantially in equilibrium, since any attempt to move it in a specific direction has been promptly neutralized.

Candlestick

charts

(this slide has been revised)Slide98

98

98

Candlestick charts

Long white or long black candlesSlide99

99

99

Candlestick charts

Long bilateral shadows

with short bodies; these

candles are referred to

as spinning topsSlide100

100

100

The equilibrium of a short bodied / long shadowed candle is very unstable, though: a strong directional movement is probably soon to come.

When

short bodies

come together with

short shadows on both sides

the interpretation is very different: they reveal a total lack of interest in that asset by the operators. Until new information come into play we can expect that asset will not move significantly.

Short bodied candles with only one long shadow are completely another matter

: if the long shadow is the upper one, for instance, it means

demand was strong for a good part of the day, then at some point supply became much stronger, pushing back down the price (see

slide

101).

Such candles are called

exhaustion signals

(they belong to the family of the so called

reversal patterns

):

demand is out

of power, and

supply is

much stronger. If a candle like this shows up on top of an uptrend it is a

warning

signal: a strong drop could be soon to come.

We added a new ingredient to the “market reading recipe”:

the specific position of the candle within the trend

.

Candlestick

charts

(this slide has been revised)Slide101

101

101

Candlestick charts

Long upper shadow, short body; on top of an up trend it is quite a dangerous signal; indeed, such candles are referred to as hanging menSlide102

102

102

Exhaustion candles can show up also in the opposite way, see slide

103,

where a circled candle shows how

a strong supply pushed

down the price

for some time duritng the day, till demand became much stronger, and pushed

back up the price. When candles like this show up on the bottom of a downtrend they mean a new uptrend

could be

at stake.

The reading of the candlestick chart, indeed, is much more complex than it might appear from this few slides, since the interpretation of each candle and group of candles depends also on the picture of the market they belong to.

The fact is that candlestick on their own are often meaningless and they do not allow to take investing decisions given of a high reliability: we need to integrate them with other indicators.

In any case we have always to remember that our trading technique needs to be effective in probability: being succesful 60-65% of the time is already a very good result.

Candlestick

charts

(this slide has been revised)Slide103

103

103

Candlestick charts

Long lower shadow, short body;

on the bottom of a downtrend it

means it is likely the time to buy.

Such candles, no matter if black

or white, are called hammersSlide104

104

104

One of the key factors in technical analysis is the degree of market participation to some relevant event, such as the generation of important price tops or bottoms.

In the field of odds, patterns that come with a huge market participation should be much more significant than those who come with a poor interest from investors: indeed, the wider the masses of investors attracted by an event, the more powerful the effect of their behaviours.

A good measure of the market interest for an asset is

the number of shares (or contracts, in the derivatives field) traded at some time in respect with the total number of outstanding shares (or the average number of traded contracts).

Key reversal patterns, such as a

hammer

or a hanging man, a price top or bottom, the golden or devil crossing of a long term average, coming with a traded volume equal to some percent point of the total market capitalization of that stock, or with a number of traded shares much higher than the daily average, are all facts to be kept in very high consideration.

Generally speaking, the higher the traded value (number of shares or contracts times their price), the stronger the power of a buy or a sell signal.

Traded volumes and valuesSlide105

105

105

Pros and cons of technical analysis and quant trading

Price charts allow traders to seek for repetitive patterns. They provide investors that trading edge he

needs

to find in order to achieve systematic profits. It is perfectly clear, though, that this result can be achieved only with a scientific approach, computer based. Knowledge of financial markets and products is the first rule, of course.

The inventory of trading strategies is so wide that a whole lifetime would not be sufficient to manually test them all. Indeed, systematic strategies can be based on just prices, prices combined with price indicators, prices combined with non-price indicators, time parameters only. A basic examples for each type:

price based: buy on any new 20-days maximum, sell on any new 5-days minimum;

price combined with some price indicator: buy on any golden cross between price and 5-days moving average, sell on devil crosses;

price combined with some non-price indicator: buy any time the number of stocks that closed

in the

green the previous day is higher than the number of stocks that closed

in the

red; sell on the opposite signal;

time based: buy every Monday morning, sell every Wednesday evening.Slide106

106

106

All the available indicators are included in trading platforms offered by banks and brokers, and in any backtesting software. Historical data are sold by banks and other companies. They can also be downloaded for free (the upload to the software for backtesting is manual, though); even open source spreadheets, such as OpenOffice, can be used for the analysis.

There is no trading strategy that cannot be tested on historical data.

What we need is to know how to do the job: what to look for and how.

A very spread – wrong – idea among quant trading’s detractors is that quant trading cannot work, because thanks to modern information technologies it is possible to inspect so many indicators, parameters and tools that sooner or later anyone can find something that works for sure; hence the trading edge cannot exist, since it could be available for anyone.

Their opinion is that results ahieved with a computer work only with paper money, not in the real world; this because the optimization of the parameters of the indicators leads to combinations so exasperated that they can work only in the paper world.

Pros and cons of technical analysis and quant tradingSlide107

107

107

Indeed, in their reasoning there are two severe conceptual mistakes. The first is that the number of indicators and tools available is so high that even with a super computer a whole lifetime would be not sufficient to test all the possible combinations; this means you cannot shoot in the dark hoping to catch something. The second, even worse, is that to use a computer in order to test all the possible combinations till we find a good one is not quant trading.

Quant trading is not a quest for the holy grail. It is not the lunatic seek for the perfect strategy. It is not about finding something that works at any cost, even if it is completely detouched from the real world.

To do quant trading means to take a good idea coming out from market observation, experience and hard work, test it on historical data in order to give it statistical consistency,

and only in the end use the computer to better it

, if it’s possible, trying different values of the parameters and filters on entries, to try and find the way to discard in advance the worst ones.

This process goes under the name of

optimization

, and the trader needs to be very careful about that. The

risk of optimization

is indeed to abuse of the computer in the spasmodic search for the perfect strategy for any market.

Pros and cons of technical analysis and quant tradingSlide108

108

The correct optimization of parameters

A typical mistake many traders make is to look only at the net total profit of a strategy: they set the values of the parameter(s) so to maximize the expected profit. What usually happens is that maximum profits show up on isolated peaks.

Maximum profits often show up with values of the parameters that make huge profits, but they are surrounded by other settings that produce much lower profits, or even losses:Slide109

109

When a profitable range is surrounded by losing – or poorly gaining – ranges, then the strategy is very dangerous: a minimum change in the market structure is enough to crumble the

sandy

castle; the ideal situation is like the one on next picture:

With such a system we can just choose the value of the parameter(s) that lie in the middle of the most profitable range: even if the market changes its structure the profit doesn’t change significantly.

The correct optimization of parametersSlide110

110

110

As Robert Pardo, the father of quant trading, teaches us, quant trading – when properly pursued – provides five major advantages in respect to discretional trading: verification, quantification, consistency, objectivity, extensibility.

Verification

: a coded trading strategy can be tested to prove its ability to gain over time; what we first want to know is

the profitability of the strategy in respect with the risk

it exposes

to

. It is very hard to keep investing with a strategy without knowing whether it really works or not; we also want to know on which markets the strategy can perform better, in which market phases (it’s better to stay away from strategies that work only in bull markets, for instance) and in which time periods (it’s

also better to

stay away from strategies that perform well only over a few time periods, and bad in all the others, even if the overall result is positive).

Quantification

: the judgement about the validity of a trading strategy goes through the valuation of several parameters;

risk-weighted return above all

, but not just it. A trading system that gains a lot of money exposing the investor to an

unbearable

risk

is not a

good

trading system. Moreover, the risk-weighted return is the only yardstick for the

comparison between different trading strategies and vehicles

.

Average trade

, number of

profitable trades Vs losing ones

,

average gain

on positive trades Vs

average loss

on negative ones,

profit factor

(total gains divided by absolute value of total losses),

highest number of positive and negative trades in a row

,

worst

historical

loss

, are only some of the key-parameters for the correct evaluation of a trading strategy.

How much money we need

to have on the account to be able to follow the strategy, even after a very bad moment (this is about money management, see forward).

Pros and cons of technical analysis and quant tradingSlide111

111

111

Consistency

: the basis of quant trading is the application of the same trading rules any time a specific set of conditions shows up. Signals to buy and sell, position sizing, risk managing, profit managing: everything has to follow a preconditioned logic

Objectivity

: a coded trading rule is not influenced by human emotions. Fear and greed, lack of confidence, hope, delusion. An automated strategy

just follows rules

.

Extensibility

: a human trader cannot follow too many markets and timeframes. A set of automated trading systems can follow all the markets in the world, 24/7. The only limit is in the computational power of the computer.

Pros and cons of technical analysis and quant tradingSlide112

112

112

How to read a system report (a short outline)

Previously we commented two parameters on the following system report: % gain and optimized pace of the average.

There are three other key parameters in that report

for the evaluation of the effectiveness of a strategy

: the number of historical trades (4

th

column), the number of profitable and losing trades (5

th

) and the ratio between the average win and the average loss (6

th

).Slide113

113

113

Net

profit and

net

percent profit

: are the two parameters unexperienced investors usually look for at first. Indeed they point out how much they can gain. Keeping just this information and discarding all the other can be a serious mistake; the first issue is that together with a profit it always comes a risk. Not knowing how much that risk is can lead to several problems. As it does not knowing how long negative series (losing trades in a row) can be, how much the strategy could lose before drawing a new equity maximum, and so on.

As we’ll see forward, in the evaluations of a trading strategy it is not just a matter of the first and the last point of the equity line (how much money

I have in the beginning,

how much money

I

could have after a while):

what mostly matters is what’s in between

. That is what tells us whether we will be able to keep following the strategy even through the hardest times or not; in other words, that line informs us

how much “pain”

we have to be prepared to stand to have the opportunity to reach some final

expected

result.

Number of trades

: this is one of the most important parameters. Since, in fact, in the field of systematic trading it is all just a matter of odds, the higher the number of historical trades, the higher the significance of the

results,

and the likelyhood of expected future gains

. Looking back

to the

system report we can state that to use the 8-days moving average provides the best cumulated gain, with 46 recurrencies, while the 6-days MA provides an expected lower return (still good) with 71 recurrencies: much more significant. The likelihood to achieve in the future results in line with those achieved in the past is greater in the second case. Some author suggest not to trust trading strategies having less than 200 recurrencies on past data…

How to read a system report (a short outline)Slide114

114

114

Number of winning Vs losing trades

: what we’ve seen so far still

is

not enough to take a decision, since, as we can see in the next column, the 8-days moving average provides 20 winning trades, out of 46 in total, and 26 losing, while the 6-days MA provides 28 winners and 43 losers. Do you think you can easily invest money knowing you are going to lose almost six times out of ten? If you can’t stand a strategy that loses 60% of the time, then discard the 6-days MA, no matter its higher statistical significance! And, by the way, how can such a strategy be profitable in the end? The answer comes from the next column.

Average win on average loss

: even a strategy that wins only 40% of the time can be profitable in the long term, since it all depends on how well it performs when it gains in respect with how bad it performs when it loses. Specificly, with the 8-days moving average the strategy gains only 43.5% of the time, but, when it

does,

it gains 3.15 times – on average – what it loses when it loses. This makes the strategy profitable in the long term. The 6-days MA performs a little worse, but still very good.

There are other parameters that come into play, of course, and to decide whether a strategy can be tradable for us is a long and complex process. Two other key parameters are

the shape of the equity line

(see forward, in the section about money management) and

the worst

historical

dip in the equity line

itself, alias

maximum historical drawdown

; these two variables can give answer to one of the main concerns we previously spoke about:

what’s in between the first and last point of the equity line?

How to read a system report (a short outline)Slide115

115

Step 3: modern trading techniques

Thomas De Mark, TDLines and TDPoints

In the middle of the ’80 of the last century, the famous american trader Thomas De Mark made deep studies about trendlines.

In his famous book “The new science of technical analysis” he tells that at some point in his career he was studying trendlines day and night with his business partner. Every day, after work, he used to print charts on paper, bring them home and draw trendlines on them using a ruler and a pencil. His partner did the same on his own.

The morning after they confronted the charts and they figured out that they had drawn different trendlines on the same charts: they did not agree in the reading of the same charts!

De Mark was really frustrated by this situation and decided to try and find a way to codify trendlines in a unique way. What he found out is that if we look at a chart left to right there are several trendlines that can be drawn, because it all depends on the starting point and the points the observer considers significant.

But the truth is that the latest information should be more important than the oldest one, and this means we have to read charts right to left!

Doing that, starting from the present day and going back, sooner or later we will find a relative minimum and a relative maximum. Then, depending on the trend we can find a second minimum or a second maximum, that joined with the first one generate a trendline. Slide116

116

Step 3: modern trading techniques

Thomas De Mark, TDLines and TDPoints

And that trendline is unique: it’s the only one that can be drawn on the chart in that specific moment. That is the only trendline that matters.

But in order to achieve that result we first need to specify how we can find the points to be linked with the TDLine.

In the interpretation of De Mark, a maximum is of some importance only when it is surrounded by two lower maximums, one

to the

left, one

to the

right (we need three bars, then to identify it); similarly, a minimum is of some importance only when it is surrounded by two higher minimums, one

to the

left, one

to the

right. De Mark calls these points TDPoints.

TDPoints can be of several magnitudes: a magnitude 3 maximum, for instance, is a maximum surrounded by six lower maximums: three

to the

left, three

to the

right. The higher the magnitude, the higher the importance of maximums and minimums.

The next step to take in order to find the correct trendlines is to understand what we have to look for in any specific situation.

Let’s look at the chart on next slide, referred to the stock A2A, weekly bars, summer 2014.Slide117

117

Step 3: modern trading techniques

Thomas De Mark, TDLines and TDPointsSlide118

118

Step 3: modern trading techniques

Thomas De Mark, TDLines and TDPoints

As we can see on the chart, the stock is in a negative trend. Remember what previously stated? When the trend is going down the only trendline that matters is the resistance.

In order to find a trendline we need two maximums, and the one

to the

left has to be higher than the one

to the

right, so that the resistance is descending.

Starting from the last bar

to the

right and moving left we find a maximum first: the circled one

to the

right.

Going left a few bars we then find a minimum (circled as well), that will help in defining the target of the trade (see forward). Going left again we reach a new maximum, higher than the first one. We found the resistance we were looking for!

That is the only trendline that matters in this trend phase. Now the question is: what do we make of this trendline?

De Mark says that when the price breaks out the resistance line (there are specific rules for intraday and end of day breakouts, but they are not important here) the price should change its direction, and it shoud also reach a specific target price (see forward).

In order to define the target of the new up trend we need some more information. First, we need to determine the exact slope of the trendline, which serves two different objectives.Slide119

119

Step 3: modern trading techniques

Thomas De Mark, TDLines and TDPoints

This is quite easy: we take the distance between the two maximums and we divide it by the number of bars between them (it’s the logic of the equation of the line linking two points).

This leads us to the average drop between the two maximums day after day: the slope of the line.

And it also allows us to precisely determine the value of the trendline every single day after the second maximum; this means that no matter the day of the breakout we can calculate the exact value of the breakout, which is the second important element for the calculation of the target price of the new up trend.

De Mark defines then three possible targets for the trends originated by a breakout:

we take the vertical distance between the lower low underneath the TDLine (the circled one on the chart) and the TDLine and we add that distance to the point of breakout;

2. we take the vertical distance between the close of the day of the lower low underneath the TDLine (the close of the day of the circled minimum) and the TDLine and we add that distance to the point of breakout;

3. we take the vertical distance between the lower close underneath the TDLine (the day after the lower

low in the chart considered)

and the TDLine and we add it to the point of breakout.

Let’s

look at some charts and try to apply this technique to them.Slide120

120

Step 3: trend indicators; the stochastic oscillator

The stochastic oscillator is an indicator that can be used to point out the strenght of a trend; it compares the distance between the closing price and the minimum of a number

n

of days to the range (the distance between minimum and maximum) of the same number

n

of

days, on a percentual basis:

Stochastic = 100 * [(Close - MINn) / (MAXn - MINn)]

MINn and MAXn are the lowest minimum and the highest maximum of the

n

days. The value of this indicator moves then between 0 and 100.

The premise is that in strong uptrends the daily close price is expected to be close to the daily maximum, then the stochastic is expected to be high; vice-versa in strong downtrends.

When the stochastic reaches specific thresholds, usually 20 on the downside and 80 on the upside, then we can say that the stock we are looking at is oversold (very weak) or overbought (very strong).

A first point of attention, then is when the oscillator reaches the bands of overbought or oversold and then gets

out (back down or back up respectively):

it means that the excess is ceasing

.

The stochastic is often combined with its moving average to improve the effectiveness of the trading signals.Slide121

121

Step 3: trend indicators; the stochastic oscillator

Usually investors use these situations to buy or sell, but they make a mistake because they are misinterpreting the indicator: the stochastic is a signal of strenght; when its values are very high and at some point they reduce it doesn’t mean the trend is changing: it just means it is taking a break, that could be only a short pause before a new explosion.

A very good use of the stochastic oscillator can be found in the book “Vivere di mercati” by Paola Gentili, one of the very few italian very successful women in this business.

According to Paola’s

studies,

the return of the stochastic out of the excess bands is not sufficient to open positions; Paola puts on her charts a stochastic oscillator and three exponential moving averages. The parameters are always the same, no matter the stock or the timeframe.

What Paola found out is that given the three following conditions:

the stochastic gets out of an overbought situation (it crosses

20 from below to above);

the three moving averages are sorted coherently with their paces;

the price crosses the first moving average;

Then the price usually moves right away to the second moving average; and if it crosses the second moving average it moves to the third.Slide122

122

Step 3: the relative strenght

One of the most interesting use of charts is to compare different assets in pairs so to inspect which of the two is stronger in terms of its trend. For example we can make the ratio between the prices of two assets at some ideal starting point, and then keep calculating the ratio day after day, so to determine which of the two assets is stronger than the other one.

Moreover it is possible to compare the performance of a stock to that of the index it belongs to, so to see whether that stock is performing like the market, over the market or under the market.

At the same time we can compare qualitatively indexes of several countries in the world, to underline the differences in terms of strenght and expectations about the future development.

It’s interesting, for example, to examine the different relative strenghts between the main european and the main american indexes; it is also interesting to look at the differences, in terms of performances, between the italian market and the french, or the german one.

What comes out of this kind of comparison is that there are countries that are already virtually out of the crisis, and other countries still embroiled in it.

One important note, though: not all the indexes are calculated in the same way, so comparisons between different markets require some kind of attention to the specific calculation methods.Slide123

123

Step 3: the wolfe wave

The wolfe wave is a pattern that shows up not very frequently on the markets, but when it does it offers very interesting investing opportunities.

Let’s see how it works. First, it is a pattern that usually takes several bars to be completed, since it is made of four key-points: two maximums and two minimums. Three of them

(points 2, 3 and 4) are very important maximums and minimums.

Let’s see the typical shape of the wolfe wave and then we’ll comment it:Slide124

124

Step 3: the wolfe wave

First, the wolfe wave is a pattern that must show up as a secondary trend in a primary trend; in simple words, the master treend has to be in a clear direction, and the wolfe wave has to show up in a correction phase (down in un uptrend, up in a downtrend).

In this case the master trend is up and the WW shows up in a downtrend of a lower magnitude.

The first point we need is number 2, then 3 and 4. Then we link 2 and 4 with a line, then we seek for a minimum 1 so that the lines 1-3 and 2-4 come together at some point in the future.

Then the market has to

make

a false breakdown of the line1-3, point

5 on the chart,

followed by a bounce that leads the price to cross the maximum of the

bar that marked point number 5

. There we buy.

Target price? The line 1-4! The longer it takes… the better it is!Slide125

125

Step 3:

Bollinger Bands

Invented by the famous american trader John Bollinger, the Bollinger Bands are a price channel used to provide a visual measure of the volatility of an asset.

Starting from the moving average of the closing price over some number of days (usually 20), subtracting and adding to it the standard deviation of daily returns over the same number of days of the moving average, multiplied by a factor (usually 2), a price channel is obtained.

The width of the bands is a function of the historical volatility of returns. Assuming that the expected future volatility will be in line with the past one, the Bollinger Bands provide an expected range for the next price movement.

Some use the BB to spot extreme price movements: when the price touches one of the bands it means a strong movement took place and a correction should come soon. This theory is based on the assumption that the price should be mean reverting, which is not.

Drawbacks: the Bollinger Bands are based on two parameters that can be optimized (the pace of the calculation and the multiplier). Changes in one or both can make the bands change significantly, giving a completely different reading of the price movement. Moreover, the forecasting power of the bands is very limited, since they can change both in terms of the average and the volatility. Still, all the days concurring to the calculation are equally weighted: a shock happened 20 days ago is weighted as it happened yesterday. And finally, as we’ll see, risk is not symmetrical in the financial markets as the BB assume it to be.Slide126

126

Step 3:

Bollinger Bands

Bollinger Bands 20-2Slide127

127

Step 3:

ADX oscillator

Some of the oscillator available in the trading platforms are designed to provide a measure of the strenght of the price of an asset.

The ADX is designed to split this information into two components: strenght and predominant direction.

On a specific time basis (usually 14 days), indeed, it is made of three lines; one points out a value between 0 and 100 pointing out the strenght of the price movement; and two others pointing out which direction of the price is prevailing.

The calculation of this oscillator is quite long, since it requires a lot of components to be combined all together in a logic sequence. By the way, it is more important to see it applied to a chart in order to understand its use.

The chart on next slide is referred to the stock Eni. The ADX is also plotted, on a 14-days basis. The black line is the ADX, the green line is the positive directional index, the red line is the negative directional index.

Readings of 30 or more of the ADX point out a very strong price movement. The two directional components then point out the prevailing direction.Slide128

128

Step 3:

ADX oscillatorSlide129

129

Step 4: risk management

Once we got access to the market, defined our approach to the market and chosen the way to analyze data and charts we still have many choices to take in order to invest money correctly.

Now let’s assume our set of rules is telling us to buy the stock ABC; now we have three main problems that have to be solved all together (we can’t solve one without considering also the other two at the same time): how much money to invest, how much we can tolerate to lose if things go wrong, where it should be reasonable to get out.

Here we start from risk management, but nothing prevents us to start from any of the other two.

Risk management in a few words: do you have a plan B? what will you do if

it turns out you’re

wrong?

Believe it or not, in cold blood it’s very easy to tell what you should do when things go wrong, but when you are looking at the chart of a stock you bought and the price is falling it is much more complicated; on one side we are aware that the price could drop even more, but on the other side our brain is programmed to be optimistic, hence the signals it sends

us

are to keep positions waiting for a bounce. The problem is that the bounce could be late, or never come.Slide130

130

Step 4: risk management

The problem is mainly psychological, but also mathematical: it originates from the incapability of many people to account for

the rule of the financial ruin: the path to recover a loss is much longer than the path to make it.

Indeed, to recover a 10% loss

it takes

a 11.1% gain! To recover a -20%

it takes a

+25%... To recover a -50%

it takes a

+100

%. To recover a -99% it takes a +9900%!!!

The higher the loss, the harder the recovery process.

Always keep in mind the relationship between profits and time: for a 100% gain you need a long time, and a long time exposes yourself to the risk of huge losses; it

might

happen, then, that in the meantime you wait to recover a loss you get a higher loss.

A very good trader usually can gain a 2-3% a month (as the algebric sum of profits and losses) on his whole available capital.

To let a single trade lose a 50% means to erase in one single shot the good job of weeks, or event months.

The bottom line is that

losses have to be managed

.Slide131

131

We cannot deal with the risk of an investment without any plan B, because it is statistically impossible for trades to be always

profitable.

There is only one way to manage losses: set a threshold (defined by us, in advance, and on the basis of some criterion statistically efficient) we are not willing to overpass; and if the price reaches it, then we get out.

Losses have two faces: the first is monetary, the second is psychological; losses demoralize us, take away our confidence in ourselves and in our strategies. The higher the loss, the worse the effect.

In the exact moment we open a trade we need to have clear in mind the plan B. It is called

stop loss

, because it is made to limit losses when things do not go as we expected.

We can re-emerge from a 10% loss; if it’s just a 5% it is even better. But we cannot re-emerge from a 99% loss.

Many of the main experts in the world suggest not to put at risk more than a 2% of our capital per trade…

Step 4: risk managementSlide132

132

Americans have a famous proverb:

“if it’s not broken, don’t fix it”

. In finance this doesn’t apply: sooner or later we have to get out from a trade, and once again, when and why is a decision that cannot be left to chances.

It might sound odd, but it often happens that

to properly manage profits is even harder than to properly manage losses

.

And again, the problem is psychological: most of the time we tend to settle for a profit that seems to be sufficient, just to be sure to have something in our pocket; if tomorrow the price keeps moving up it doesn’t matter.

The idea is that it is better an egg today than a chicken tomorrow, because the egg today is sure, the chicken tomorrow is not.

The common denominator for investors who settle for the egg today is “what if tomorrow the price drops?”

Actually the problem is usually of a different nature: they have no idea about the effectiveness of their strategies.

Step 5: profit managementSlide133

133

Without a test of a strategy it is not possible to form expectations in terms of its potential profits. How can we tell it’s better to get out of a trade after a 5% profit or a 10% profit, if we do not test the two alternatives? These information can come only from a backtest of the strategy, made on a data series long enough to give the system a statistical consistence and significance.

It is all a matter of coding a strategy so to be able to simulate its behaviour on past data, combining it with an exit strategy of some sort.

For instance,

buy when the price

crosses from below to above

its 5-days moving average, sell when the price reaches a

new 20-days maximum

.

How much money can be made with this strategy?

How much risk does it expose us to? Only

a backtest can give answer to this question.

One of the most popular proverbs in the field of trading is “cut losses and let profits run”. When do we have to cut losses (plan B)? Till when do we have to let profits run? Only a backtest can answer these questions.

Step 5: profit managementSlide134

134

Money management is by far the most important issue in the management of an investment, and still one of the most ignored by private investors.

The problem here is to try and find the best allocation of capital among all the available investment vehicles. Here another famous american proberb can help:

“do not put all your eggs in the same basket”.

Because if the basket falls down, all the eggs brake at the same time. Talking about investments, the

invitation

is to not put all your money on a single instrument, but to split it into several.

If we have a million to invest, instead of buying

one million

euro of

a single stock maybe

it would be better to buy a quarter of a million of government bonds, a quarter of a million of corporate bonds, and

invest the

remainder half a million in shares. And about the last half a million,

maybe a

hundred thousand in the stock 1, another hundred in the stock 2, and so on.

Doing this we create a portfolio of instruments given of some degree of expected risk and some expectation in terms of return. It’s the process

known as

diversification

.

Step 6: money managementSlide135

135

The idea si that a single company could go bankrupt; it’s a matter of

specific risk

, that is, the risk that a specific company fails; if all my money is invested on that company… disaster!

Generally speaking, the probability of bankruptcy of a listed company is very low, but the probability of the simultaneous bankruptcy of twenty listed companies is zero; that is the reason why it is better to diversify: because the probability of a combination of tens of unfavourable events all at the same time is not conceivable. The best diversification is given by the whole market: it can be proven mathematically that the risk of the whole market –

systematic risk

– is the least risk we can reach thanks to diversification.

Apart from the worst case scenario (bankruptcy), the probability for a single company to see a dramatic drop in price is much higher than the probability for that to happen on ten stocks at the same time, unless something very bad happens on the whole market (think of the twin tower attacks in the year 2001, or the bankruptcy of Lehman

Brothers in 2008).

Let’s look back to the idea of the diversification seen on previous slide. An investor with a low inclination to risk could object that a half of the capital on stocks is too risky, and that would be much better to invest half a million in government bonds, 300k in government bonds, and only 200k in stocks. The point is that

there is not a better choice

no matter what

,

because it all depends on the personal inclination to risk of the investor.

Remember what previously stated:

it is all just about risk and return

.

Step 6: money managementSlide136

136

Let’s

go back to the nature of the problem: the best allocation of capital. When the investor has decided how much capital has to be allocated to the stock market, there is still a problem to face: how much capital on each stock?

Let’s consider a simple case: we have one hundred

thousand euro

to invest in stocks; suppose we have a systematic trading strategy that tells us every day what to buy and sell. Today the strategy tells us to buy Eni, Telecom Italia, Fiat and Unicredit. How much money do we invest in each

?

Here we have basicly two issues: theoretically speaking, in the moment we open the trades all of them have the same risk, since nobody can tell us today which of them will return a gain and which a loss.

As a consequence, at least theoretically we should invest the same money amount on each. This is a fixed capital money management: same money in any trade.

The second issue is: what money amount in each trade? 25000?

It sounds reasonable, at a first sight, but what if tomorrow the system tells us to buy Banca Intesa? We do not have money!

What if all the first four trades are loosers and Banca Intesa was the only one profitable?

Step 6: money managementSlide137

137

To get out of this problem we need to know the maximum number of simultaneaous trades the system would have had in the past, and assume that history could repeat itself in the future.

Now suppose, that our system opened at most ten trades simultaneously in the past. Can we then invest ten thousand euro per trade? Unfortunately it could not be the correct value. First objection: what if from now on one day the system tells us to buy eleven stocks? The future should be similar to the past, not identical

!

Second problem: imagine that today the system tells us to open ten trades; the whole capital is invested.

Imagine, then, that tomorrow all trades are closed, with a net global negative result (the algebric sum of profits and losses), let’s say -10000 euro.

Now the available equity is 90000. Tomorrow the system tells us again to buy ten stocks: we cannot invest 10000 each anymore, because we don’t have enough money. So we have to invest at most 9000 each. A system born as fixed capital became fixed fractional: in each new trade we invest the same fraction (a 10% here) of the available equity.

Step 6: money managementSlide138

138

This

may not be a problem: sometimes the fixed fractional method produces better results than the fixed amount one. But that is not the

point.

The

point is that depending on the choices we take in terms of money management change the results of our trading, since to gain or lose a specific percentage on

90000

euro

is

not the same than to gain or lose the same percentage on 100000 euro

.

Unfortunately, things are much more complicated than that, and it’s time to understand why money management, risk management and profit management are so linked together.

How much we gain or lose in a single trade depends not only on the entry strategy (what to buy and why), but also on the management of losses and profits.

This explains why to invest money is not as simple as many peope think. Because the scientific test of the effectiveness of a strategy is strictly related to all the aspects of the problem, hence it depends on the entry strategy, but also on the stop loss and the take profit policies and even on how we allocate capital among all the products.

Step 6: money managementSlide139

139

Think

of this: if any time you buy you set a 1% stop loss from the entry price, you know that in any trade you put at risk a low money amount, but you can expect to be stopped out many times, because the stop loss level is

too

close to the entry point; vice-versa, if you put a stop loss at a -10% from your entry point, then you can expect that you will be stopped quite rarely, but any time you get stopped you will lose much money.

Which

one is the best? No one can tell without a test, because it depends on too many variables.

Now think of profits: if you set a 1% target profit, you will very likely gain most of the time, but only a 1%; if you set a 10% target profit, you will gain rarely, but much more each time. Again, only a test can tell you which one is the best choice.

Step 6: money managementSlide140

140

In order to make things clearer in terms of the impact of the management policy on the financial result of a trading strategy, now i will show you the results of the application of four different techniques

to the

same strategy.

The strategy is the most simple one: crossings between the close price and a 20-days moving average. The stock is Azimut, listed on Milan stock exchange. No stop loss nor take profit policies are applied: the system buys on a golden cross and sells on a devil cross.

If we record all the trades that could have been made between January 2000 and the end of September 2013 in a table sorting them chronologically, reporting the profit or loss of each and calculating the cumulated results, we can get to the equity line of the strategy, that is, starting from the initial capital, what happens all the way long, till the last day.

Since, in this case, entry, stop loss and profit management rules are always the same, the only thing that can change the results of the strategy is the way we use money. About that, suppose to have an initial capital equal to 20000 euro.

We’ll now see what happens applying the following different

money management policies

: fixed capital, martingale, anti-martingale, fixed fractional.

Step 6: money management applied to stocksSlide141

141

Here how the four techniques work:

1. Fixed capital: for any new trading signal we invest always the same money amount, 10000 euro. Just a half of the available money because we need to keep a reserve of money in case the first trades were negative. Here the equity line chart:

Step 6: money management applied to stocksSlide142

142

2. Martingale: on the first trade we invest 10000 euro; if the first trade is winning, on the next one we invest a fixed percentage less (5% less in this case); vice-versa, if the first trade is losing, then on the next

trade

we invest a

fixed percentage more

(5% again). And so on. The martingale comes from

the field of gamblers

, typically black jack players. Equity line:

Step 6: money management applied to stocksSlide143

143

3. Anti-martingale: on the first trade we invest 10000 euro; if the first trade is losing, on the next one we invest a fixed percentage less (5% less in this case); vice-versa, if the first trade is winning, then on the next rade we invest a fixed percentage more (5% again). And so on. The

anti-martingale

comes from

the idea that in trading each trade is independent from the previous, and that usually positive and negative trades show up in sequences.

Equity line:

Step 6: money management applied to stocksSlide144

144

4. Fixed fractional: on any trade we invest always the same fraction of the available money (20000 euro), the 50% in this case. On the first trade we then invest 10000. Then being the first trade winning, the capital will increase. Assume to gain 1000 euro on the first trade: the capital is now 21000, and it means that on the next trade we invest 10500 (always a 50% of the available money); vice-versa if the first trade is losing. This policy resembles an anti-martingale, and is just a capitalization of a half of previous profits or losses. Equity line:

Step 6: money management applied to stocksSlide145

145

Step 6: a full view on the four MM techniques

Fixed capital

Martingale

Anti-martingale

Fixed fractionalSlide146

146

Step 6: what equity lines are for

The equity line

provides

several key-parameters for the valuation of a trading strategy. First, the maximum historical loss between two consecutive growing peaks of profit. That is the so called

maximum drawdown

, and is a first measure of risk of a strategy.

A paramount concept in trading is that

it not only matters where you go, but moreover how you get there

. A strategy

that brings

exceptional profits exposing the trader to

unbearable risks

is not a good trading strategy.

In looking at an equity line chart, unexperienced investors usually just look at two points: the first and the last. What’s in the middle is not their concern. And they make a huge mistake, because

what happens in between is a measure of the capability of an investor to stay still in front of events without intervening on the strategy even when it loses money

.

Look back to the fixed fractional chart (previous slide, bottom right): it gains about 36000 euro (because from the final value, 56000, we have to cut the initial equity, 20000), much more than any other money management policy. But it exposes the trader to a risk to lose around 11000 euro, much more than any other trading strategy. How can we decide which technique is better? We have to compare them under the same meter: the so called

profit factor

.

The profit factor is the fraction between the sum of all the gains of positive trades and the sum of all the absolute values of the losses of negative trades. A value over 1.3-1.4 points out to a good trading strategy. The higher, the better, of course.Slide147

147

Step 6: what equity lines are for

Moreover, the maximum drawdown gives not just a measure of the pain the investor has to suffer to reach the final profit: it also tells him something about

how long can hard times last

.

Look back to the chart of the fixed fractional method. The maximum drawdown lasts from the trade number 28 to the trade number 55, and till the trade 58 there are no new gains.

Some investors cannot stand 30 trades in a row not bringing new cash on their accounts. The equity line can then tell them whether the strategy might suit them or not.

And yet there is another issue to face: one of the main problems in the maximum drawdown is that it makes many investors

abandon

their trading because they can’t stand losses on the long term. Many money management experts suggest that in passing from a simulated trading to a real time trading it would be better to allocate a money amount equal to

two times the maximum historical drawdown

, so that global losses do not incide too much on the investor’s capital and his capability to keep investing with the strategy. This is also valid considering that the worst past case is not necessarily worse than the future worst case:

the future could bring losses higher than the historical ones

.

The problem is that

the more money we allocate to the strategy

, to feel more comfortable,

the lower is the return

on the initial equity… It is again a problem of equilibrium: we need to start with a capital sufficient to be not afraid of temporary losses and to be able to keep trading after a hard time, but if we allocate to the strategy too much money the return can be not sufficient to justify its use.Slide148

148

All the steps together: a full strategy in details

A famous american trader named Jeff Cooper wrote a book in the early 2000s, titled “hit and run trading”. As the title says, it’s a collection of strategies for the very short term (1 to 5 days).

One of the most common strategies is called 1-2-3-4 and basically it looks for a short correction – 3 days – in a strong up movement, followed by a bar indicating that the trend is moving again in its original direction. This signal is given by the breakout of the maximum of the third candle of the correction.

Cooper says he takes profit usually within a couple of days, depending on his

experience, feelings and sensibility.

This behaviour cannot be copied by any other investor, since no one else can read Cooper’s mind, so if we are willing to use Cooper’s technique we need to find a way to manage positions that suits our personal nature.

Here follows a study of several approaches that can be followed.

Stop

loss

policy:

we get out if the price breaks out on the downside the minimum

of the 3 days correction considered for the buy setup. The

money management is fixed capital: 5000 euro per

trade.

Assets under analysis: italian high, mid and small capitalization stocks.Slide149

149

All the steps together: a full strategy in details

Eight different profit management policies have been tested:

1. take profit when it’s 1:1 with risk

2. take profit when it’s 1.5:1 with risk

3. take profit when it’s 2:1 with risk

4. manage the position dynamically using as stop profit the minimum over the past three days

5. get out the second day on close

6. get out the third day on close

7. get out the fourth day on close

8. get out the fifth day on close

The analysis showed what follows:

1. on high capitalization stocks the strategy doesn’t work: profits are too low in any case, and many profit management policies lose money

2. on mid and small caps the stragegy seems to work (only 24 stocks have been tested,

and 204 trades in total have been analysed)

3. the best profit management policy in any case

has proven to be number

4: dynamic management on a 3-days

basis

But pay attention: this applies only to this strategy, for the italian stock market and for the timeframe observed (daily). Nothing can tell you it could work the same way on other stock markets, in different timeframes.Slide150

150

Step 7: the choice of the instrument

Here a mere list of the instruments we can buy and sell either on our own or with the help of a dealer:

stocks

corporate bonds, convertible bonds, warrants

government bonds

mutual investment funds, hedge funds

ETFs, ETCs, ETNs (generally speaking, ETPs)

futures contracts

option contracts

covered warrants

certificates

binary options

turbo contracts

contracts for difference (CFD)

foreign currencies (Forex)

Whichever the strategy that we choose, knowing each asset class can lead to maximize our gains, minimizing our risk. Knowledge of the available instruments is very important.Slide151

151

Step 7: the choice for the instrument

Let’s now assume we have a buy signal on the stock ABC. How do we follow it?

This might seem a trivial question, but it’s not. Indeed, till about 15 years ago we would have had

no other choice but to buy shares

of ABC,

paying their entire value

on purchase.

Today the world is different; first because to buy the stock ABC we can have several different instruments (shares, certificates, CFDs, futures, options), each given of a specific risk-reward profile, which we have to fully understand in order to take the best decision; second, even if we decide to buy shares we don’t always have to pay their whole value, since the trade could be opened using only a fraction of its value.

The choice to take, then, is on two different levels:

what instrument

to use and whether to use a

financial leverage

or not.Slide152

152

Step 7: the leverage

The leverage is an artifice that provides interesting opportunities for many investors, especially those given of a very small money amount to invest.

The idea is quite simple: given a 10000 euro capital and a 10:1 (ten to one) leverage, we can open positions for up to 100000 euro, ten times our money.

Think of a trade that restitutes a 10% profit on ten times the money we have: we double our real money!

The problem is if things go the other way: a -10% on a capital 10 times bigger than my real money… no more money!

Unfortunately, the leverage works in both directions

.

Many investors think that thanks to the leverage they can become traders with a small money amount on their accounts. Unfortunately it is not possibile, and to prove that we just need to look back to one of the equity lines seen on slide

145.

Let’s focus on the fixed amount: 10000 euro per trade.

Here somone could say: “i have a 10:1 leverage, so i dont’ need 10000 euro on the account, since 1000 are enough”. This is wrong, because the leverage can reduce the need for capital per trade,

but it doesn’t change the maximum drawdown

of the strategy, that is always 5000 euro and something. We cannot stand a 5000 euro loss with only 1000 euro on our account!Slide153

153

Step 7: the leverage

Moreover, many of the greatest experts in the world in the field of money management suggest to put at risk not more than a 1-2% of our capital per trade.

Looking back to the equity lines on slide

145 we

can make a rough estimate of the worst single loss ever seen: trade number 31 (fixed amount chart) loses about 1000 euro.

If the loss per trade has to be at most a 2% of the available money… then the money available has to be 50000 euro! And it rises to 100000 if we put at risk only a 1% per trade.

Given a 10:1 leverage we still need something between 5000 and 10000 euro to be able to follow the strategy even in the bad times. If we double the maximum drawdown as the esperts suggest, we need to double the available money as well.Slide154

154

Step 7: costs of the leverage

Most of all, the leverage is not for free.

If we have 10000 euro on our account and we ask the bank to give us a 10:1 leverage we are implicitly borrowing 90000 euro.

On the money lended the bank wants to be paid interests, which usually account for the yearly market interest rate (very close to 0 today) plus 6-8 percentage points.

Let’s do some quick accounts: we buy 5000 shares of a stock quoting 20 euro per share, for a total value of 100000 euro, using a 10:1 leverage; this means that 10000 euro are ours, the rest is borrowed. On this 90000 euro borrowed the bank applies an average yearly rate of interest equal to 7%. On a daily basis we then pay 90000*7%*1/365=17.26 euro.

Keeping the trade opened three months we pay more than 1550 euro in interests!Slide155

155

Step 7: short selling

Short selling

(or just

short

),

vendita allo scoperto

in italian, is another artifice of the modern financial creativity and is a way to “bet” on the downside, on a drop in the price of an asset. It’s the opposite of the so called

long

trading (to buy under the expectation of a growth).

Why could it be reasonable to invest on the downside? Another famous american proverb states that

the markets go up with stairs and down with the elevator

: the average speed of up moves is generally lower than the average speed of down moves.

This comes from the market psychology: in downtrends investors tend to runaway because they are afraid of remaining stuck in a losing trade, sometimes generating the so called

panic selling

; investors sell anything for any price just to be sure to get out, often out of any logic.

An example above all: between march 2003 and may 2007 the italian stock index FTSE Mib lived one of its best periods, raising from 20,324 to 44,363 points; about 24,000 points in about 50 months. Then, from May 2007 to March 2009, because of the global crisis triggered by the sub-prime mortgages it dropped down to 12,332 points, about 32,000 less in 22 months. Less than a half time to lose much more in terms of value.

This is panic selling: stocks are sold for prices always lower, often way under their real value, just to convert a position in shares into a liquid account. “

Cash is king!”

say the americans. No money, no good deals; better lose money today to have the money to gain tomorrow.Slide156

156

Step 7: short selling (of stocks)

The idea of the short selling is then to sell something at some price trying to buy it back for a lower price after a while. But how does it work?

To sell short means to sell something we do not have. If the long buyer has to be afraid of price moves on the downside and gains from price moves on the upside, for the short seller it’s the opposite: he has to be afraid of an up movement, and gains from a drop.

How can we sell something we do not have? Of course we have to borrow it: we borrow a stock, we sell it for a price, let’s say 10, so we get paid 10 per share. After a while the price drops to 9, so we buy back the stock (buy to cover), pay 9, and we restitute it to

its

owner. The gain is 1 per share.

How is all of this possible? The market needs to define a set of rules to follow and to find the way to manage all the process. This means that short selling is possible, but it costs. Indeed, usually banks ask for a fixed cost for the procedure (about 10 euro); then they ask for an interest on the value of the position. This for a clear reason: in the moment we sell something that doesn’t belong to us we get paid money we don’t own. We could take that money and use it for other investments, for example to buy a risk free government bond. This would allow us to gain on money that doesn’t even belong to us! The interest we pay on short selling are the same we pay on the leverage.

An important note: the costs of short selling apply only to stock or other assets that need to be borrowed to be sold short. For other assets – derivatives, for instance – short selling is much easier.Slide157

157

Step 7: short selling (of stocks)

Summing up, there are 4 different types of trades:

to open a long: to buy something assuming its price will increase;

to sell: to sell to close a long;

to sell short: to borrow and sell something assuming its price will drop;

to buy to cover: to buy to close a short.

A buyer can then be someone who buys out of the blue, opening a position that will profit if the price increases, or

someone

who is closing a short position, to end it; a seller can be someone who closes a long position to end it, or someone who is opening a short position, to gain from a price drop.

In some peculiar situations short selling can be inhibited by the market authorities in order to prevent too speculative behaviours in front of serious situations on the markets (terroristic attacks, strong natural phenomena,

shocks of some sort).

Blocks to short selling are justified when panic selling shows up, because in these situations short selling could seriuosly worsen the drop in the prices.

One last note: it might sound strange, but to short sell we need to have money on our account as a proof that we will be able to pay in case of losses. This money is called

margin

and it is generally equal to a half of the

value

of the position. Moreover, the money we collect from short selling cannot be used for other investments, since it is stuck in the so called

margin account

.Slide158

158

Step 8: how to physically buy and sell

Now suppose that we have a full working trading system, which now is telling us to buy the stock ABC only if the price crosses a specific threshold; the order has to be considered valid till today’s close, and if it is not executed at that time has to be cancelled.

How can we do all of this?

We open our trading platform, look for the stock ABC, open its

negotiation book

(or just book) and place the order.

What is the book? How can we put in place an order? Which

parameters

can be set? Can we decide how long (in terms of time) make it be valid? Can we cancel it if we change our mind? Modify it? Do we have to pay when we modify it? How can we know if our order has been filled?

Basicly, then, we now have two issues to deal with:

how to send orders

to the markets and

what kind of orders

we can send.

As previously stated, a trading day starts with an opening auction, then there is a continuous phase and finally a closing auction.

The element in common in all the phases is the book. It’s an electronic table that shows all the buy and sell proposals sent to the markets from operators at some time. All the proposals are collected by the systems of the stock

exchange

, grouped and sorted according to some rules, which we’ll see.Slide159

159

Step 8: how to physically buy and sell

There are two important differences in the dynamics of the auctions and those of the continuous phase;

first

, during the auction there are no trades until its end, while in the continuous there are very frequent trades (generally); this doesn’t mean there are no movements in the auctions; to the contrary, proposals are sent, modified and cancelled continuously, as during the continuous phase, but nothing is traded till the end of the auction. This means that in the auction phase operators can also send to the markets orders to buy or sell for prices out of any logic, since there is no risk to be executed.

Second

, the prices of the orders sent during the auctions are not necessarily the prices operators are seriously willing to buy or sell, since the peculiar feature of the auctions is that, at their end, anything can be traded is traded at the final price of the auction; this means that a buy order, if filled, is filled at the auction price, not at the price offered by the buyer. At the same time, a sell order, if filled, will not be filled at the price asked by the seller, but at the auction price. The key element here is that

the auction price cannot be detrimental

in respect with the prices offered: who is executed is executed at a price

equal or better

than the one offered.

Puzzling? Don’t worry, everything will be clear soon!Slide160

160

PART

2

HOW STOCKS ARE TRADED:

Negotiation book, opening and closing auctions, orders, liquidity, risks, opportunities.Slide161

161

The opening auction

The opening auction phase on the italian stock market lasts from 8 to 9 am and is divided in three phases: pre-auction, validation, open.

A curious element is that the end of the acution is not set at a specific moment: it happens

randomly

in any second of the minute from 9 to 9:00:59 am. This so called random minute has been introduced to reduce the risk of market manipulations oriented to conditionate the opening price. In order to understand this issue we first need to see many other elements, most of all the books, and how they can be used to manipulate the price (committing a phelony, by the way!).

Let’s go back to the pre-auction phase. Here the IT systems of the stock exchange collect all the orders coming from the operators, group and sort them according to some rules: all orders to buy or sell for the same price are grouped so to show a single total quantity to buy or sell. Still, the system keeps a trace of the exact time of each order (we’ll see why very soon).

The system, then, shows a synthetic data in the book, but is always perfectly aware of the maximum detail: for each order it keeps trace of the operator who sent it to the market, the quantity offered or asked, and the precise time of

appearance.Slide162

162

Opening auction

If we open the auction book of any stock

at any

moment of the pre-auction phase and we take a picture of it, it will be more or less like this:

As we can see, the book is a symmetrical table. On the left there is the

bid

,

denaro

in italian, the buyers; on the right there is the

ask

,

lettera

in italian, the sellers.

Every level shows the

cumulated quantity

willing to be bought or sold for a specific price. But, as previously stated, the system keeps a trace of all the exact moments the single orders have been sent to the

market (see the right side of the picture for an example of a splitted level).

This because as we’ll see the time priority here is fundamental.

Buyers are sorted top to bottom in respect with the prices offered to buy: doing so, in the first row there are

the best buyers from the point of view of the sellers

.

2000 8:01:34 a.m.

1000 8:02:15 a.m.

1000 8:05:52 a.m.Slide163

163

Opening auction

Indeed, if you are willing to sell something (anything) in a

market

where all

the

buyers are clearly visible and offer public prices, then you will

sell

to who is willing to pay more!

Here the best buyers are willing to pay 7.65 euro per share; but as previously stated this is not necessarily the real price they are willing to pay: to the contrary, it happens that they could offered such a high price (much higher than any other price in the book)

just to be reasonably sure to be in the first line of the book.

Why do they do so? Because the logic of the execution of orders at the end of the acution, just like in the continuous phase and the closing auction, is

first in, first out

: who comes first and offers a higher price has the right to be executed first.

Looking back to the book on previous slide, if after a while some new buyers come into play offering 7.70 euro per share they become the new best buyers, overpassing the previous ones and making all the lines in the bid side of the book

to shift

down one position.

The only way to be sure not to be overpassed by anyone is to send an order to buy with no limits in price, that is, a so called

market order

,

ordine al meglio

in italian. To be stringent, the definition of market order applies to the continuous phase; in the auctions this kind of order is called

at auction price

: the buyer will buy at the final price of the auction, the seller will sell at the final price of the auction. These orders are always in the first line and in the price field there is a score. We will spend a lot of words on order types forward.Slide164

164

Opening auction

The difference between sending an order to buy at auction price or sending a buy order for a high price, high enough to be in the first line of the book, is that in the second case we face the risk to be executed at most for the price we offered, in the first we face the rislk to pay

a price

way higher then what we intended to do. It all depends on how the auction ends.

Let’s take a step back. On the right hand side of the book there are the sellers, sorted bottom to top in respect with the prices; doing so, in the first line there are

the best sellers from the point of view of the buyers

: imagine to be at a public fair where all the merchants sell the same good, cleraly showing their prices; you will go buy from the merchant asking for the lowest price!

On the first line of the book we have then the best buyers and the best sellers, then in the second line there are the second best buyers and the second best sellers, and so on. The standard book shows five lines, but today books with a deeper view on the markets are

available (up to 20 levels).

Time to see how the auction

works.

Buy

and sell proposals are sent, modified and cancelled all the time, and the system updates in real time the price that would be the opening price if the auction closed in that exact moment: that is the so called

theoretical opening price

. Slide165

165

Opening auction

Here we may begin to understand how the auction book can be used to manipulate the opening price if there were no market procedures to try and prevent it: each new, modified or cancelled order impacts on the real time theoretical opening price.

But at some time – in the random minute (see previous slides) - orders become definitive: who’s in is in, who’s out is out. No more apportunities to change or cancel orders or to send new ones. And if we cannot tell when the auction will end it can be very risky to try and manipulate the market with fake orders, because a fake order sent a fraction of a second before the closing of the auction becomes a real order!

We will see a fully detailed attempt of market manipulation forward.

At any time during the auction the systems calculate the theoretical opening price

applying four basic rules sequentially

:

the opening price is the price that maximizes the tradable quantity;

if two or more prices maximize the tradable quantity, then the opening price is the price that minimizes the so called

imbalance

: what is left out;

if two or more prices maximize the tradable quantity and minimize the imbalance, then the

opening

price is the one closest to the previous day’s close, here called

control price

;

if two (here there cannot be more than two) different prices maximize the tradable quantity, minimize the imbalance and are at the same distance

from

the previous day’s close, then the theoretical opening autcion price is the highest

of the two.Slide166

166

Opening auction

The four rules are sequential

, meaning that each

one after

the first

has to be applied

only if the previous one couldn’t

lead to a single

opening price. With the fourth rule it is mathematically impossibile not to come to an opening price.

And still it is a theoretical opening autcion price

, since to become the opening price it has to pass a final test: the validation (see forward).

How does the system determine the theoretical opening price according to all the proposals in the book? To answer that we need to look at the book from a different perspective,

rearranging

data in a different way.

First we build a new table, made of five columns

.

In the first column we place all the prices that appear

in the

book

– only

the visible

ones – sorted

bottom to top.

In the second column we place the quantity that can be filled for each of the prices in column one on the side of the buyers; in the third column we do the same on the side of the sellers; then we compare the quantities that can be filled on both sides (fourth column).

The fifth column is for the

imbalance

, that is needed if we have to procede to the second tule to find the theoretical opening price. We can leave it blank, for now. Look at the picture on next slide.Slide167

167

Opening auction

Where

do the quantities in the picture above come from? Any buyer that comes into play usually has a clear idea about the limit price he his willing to pay; this limit, though, applies only on the upside: it’s a

lower or at most equal

condition; “i don’t want to spend more than X”. But if we can have what we want paying for it a price lower than the one we thought… much better!

Since the most conservative buyers here are offering 7.37 per share, any price equal or lower than 7.37 will suit them, like anyone else who initially offered any other price higher than

that;

in other words, each price equal or lower than 7.37 suits all the buyers that appear

in

the book, hence the quantity that can be bought is the sum of all the quantities on the bid side of the book,

16,000

shares in total

.Slide168

168

Opening auction

If the price increases to 7.38 all the quantities asked for any price equal or higher than that can still be filled, but who was willing to pay 7.37 will be cut out: that’s why only

11,000

shares can be filled this time.

If the price moves

up to

7,40 the buyers offering 7.38 are cut out, and the quantity that can be traded drops to

8,500

shares; and so on, goinig up with the

prices

till the highest one, 7.65, for which only

1,000

shares can be traded.

Now let’s reverse the reasoning, and see

what happens on the sellers’ side

.

Who’s willing to sell for 7.24 euro per share will never sell for any price lower than that, but could be very glad to sell for any higher price, if possible; that’s the reason why on the ask side the tradable quantities increase together with the prices: for 7.24 can be sold

4,000

shares, for 7.32 we catch also the second best sellers and the quantity raises to

6,800

, for 7.42 it’s

8,000

, and so on, up to

17,500

shares for any price equal or higher than 7.46.

Now let’s move by rows

. For the price 7.24 there are buyers willing to buy up to

16,000

shares and sellers willing to sell at most

4,000

: that is the maximum tradable quantity for that price. Generally speaking, the tradable quantity in each row is the minimum between the two quantities on the bid and ask sides.

For 7.32, then, we still have

16,000

shares on the bid side,

6,800

on the ask side, then the tradabel quantity is

6,800

. And so on. The result is shown in slide

169.Slide169

169

Opening auction

Now the system is able to check whether the first rule is enough for the determination of a theoretical opening auction price. Unfortunately there are four different prices that satisfy the first rule: 7.32, 7.37, 7.38, 7.40.

We need to try the second rule

, then: among

only those four prices

we look for a unique price – if it exists – that minimizes the untradable one, the imbalance.

The imbalance is simply the absolute value of the difference between the bid and the ask quantities: that is, indeed,

what remains out

after the tradable quantity has been traded. The following picture shows the imbalances for the four prices that came out from the first rule. Slide170

170

Opening auction

As we can see in the picture on previous slide, 7.40 maximizes the tradable quantity and minimizes the imbalance at the same time:

that is the theoretical opening price here

.

This is how it works on any listed stock: each stock has its own book and its own opening auction, that works

this

way. Once the theoretical opening price is calculated there is a final test that has to be passed:

validation

. The opening price has to be within a plus or minus 10%

(5% in some cases) from

the previous day’s close; if it’s out of that range another auction takes place (see forward).

Now let’s see another example, that requires to go on till the fourth rule. Suppose the book at the end of the auction is the following:Slide171

171

Opening auction

Neither the best buyers, nor the best sellers specified a price, and that’s why in the first line there are two scores. Let’s see what happens applying rules number one and two:

Both 8.40 and 8.42 maximize the tradable quantity and minimize the imbalance, hence we have to

move

on to rule number three: the opening price is the one closest to previous day’s close. Imagine that price is 8.41. The two prices are at the same distance from it, then we need to move on to the fourth rule: the opening price is the highest of the two, 8.42.

This price passes also the validation phase, since it is set just a cent from the control price.

In case the price didn’t pass the validation phase, a so called

volatility auction

would take place. It follows the same rules of the standard auction, but lasts less; it is also repeated until the market comes to a price inside the range set by the stock exchange.Slide172

172

Opening auction

In limit situations the stock exchange can decide to enlarge the range for the validation. This happens when extreme price movement during the auctions do not come out from speculative pressures but from specific market events. Look, for instance, at the chart of Saipem during the year 2013!Slide173

173

Opening auction

Immediately after the end of the opening auction the continuous phase takes place, and the book moves onto the next phase showing what remained out at the end of the auction. Indeed, at the end of the auction all the trades that can be made are made, all at the same price: the opening price.

Let’s go back to the example on slides

169-170.

The opening price is 8.42 and it makes possible to trade

3,000

shares.

In order to understand how the book remains after the auction we have to go back to the original one and cancel one by one all the trades that can be made, starting from the first line and going down till we reach

3,000

shares. On the bid side, then, we take away the 400 shares for the opening auction price, then the 600 for 8.64, then the 800 for 8.43, and finally the

1,200

for 8.42.

All these shares are traded for 8.42 euro each

. The last ones are then those who are paid exactly the price offered; in all other cases the final price

is better

than the one

initially offered

.

What is left out is only the

1,200

shares for 8.40 euro each, that are transferred

entirely to

the continuous phase as they are.

On the sellers’ side we take away the first 400 shares for the opening auction price, then the 600 for 8.24 and the

1,200

for 8.40; and again all these shares are sold for 8.42. Look what is left out: the

1,200

shares for 8.42, the exact opening auction price. How odd! Indeed it happens… And it happens because those buyers come after those willing to sell for something less. The last line of the ask side is then transferred to the continuous phase as it is.Slide174

174

Exercises on opening auctions

Given the following opening auction book

Find the opening auction price and draw the resulting book after the auction.

Given the following opening auction book

Find the opening auction price and draw the resulting book after the auction.

BID Q

BID P

ASK P

ASK Q

1000

-

9.50

200

500

9.70

9.60

400

700

9.60

9.80

600

1300

9.50

BID Q

BID P

ASK P

ASK Q

1000

8.45

-

450

200

8.44

8.43

750

700

8.43

8.44

600

1000

8.42

8.45

200Slide175

175

Exercises on opening auctions

Given the following opening auction book

Find the opening auction price, given a previous day’s reference price of 6.735. Then draw the resulting book after the auction.

Given the following opening auction book

Find the opening auction price and draw the resulting book after the auction.

BID Q

BID P

ASK P

ASK Q

3000

-

-

1000

2000

6.75

6.70

1000

4000

6.72

6.71

1000

1000

6.71

6.73

7000

12000

6.70

6.75

7000

BID Q

BID P

ASK P

ASK Q

1000

-

-

200

500

10.00

9.70

400

700

9.90

9.80

600

1300

9.80

9.90

800

2000

9.70

10.00

1000Slide176

176

Exercises on opening auctions

Given the following opening auction book

Find the opening auction price and draw the resulting book after the auction.

Given the following opening auction book

Find the opening auction price and draw the resulting book after the auction.

BID Q

BID P

ASK P

ASK Q

3500

-

-

6800

1700

17.0

16.4

1400

5000

16.9

16.5

1200

3200

16.7

16.8

9600

3800

16.6

16.9

6000

BID Q

BID P

ASK P

ASK Q

1500

-

-

2250

3750

15.77

15.74

1400

3150

15.76

15.75

4750

2500

15.75

15.76

2250

3000

15.73

15.78

3000Slide177

177

Random

minute and market manipulation

As previously stated, the randm minute has been introduced in order to prevent market manipulations attempting to change the opening price. It often happens, indeed, that many proposals to buy and sell are sent with prices set up to condition the opening price. If the exact ending time of the auction was known, then it would be possible to manipulate the opening price till a second before the closing of the auction, and then cancel the fake orders so to be not executed, obtaining a manipulated price in any case.

Why should people try to manipulate the opening price? To understand that we have to remember that prices move up and down according to expectations, and expectations are influenced by the news and parameters that can be observed on the markets. When a stock moves up it attracts buyers. It’s a psychological factor: many investors are afraid of being cut out from a good deal or to remain stuck in bad deals, and that’s why when prices raise many investors are mentally pushed to buy and when prices fall many investors runaway from stocks selling for any price just to get out.

Speculators usually monitor the best an the worse stocks every day, because generally they are the stocks to trade to make good gains over short time periods; then imitators usually do not wait to come into play and give more fuel to price movements.

If we have an open trade on a stock and we wish to push up the price, then, we can try to manipulate the opening price: a good start will attract new buyers, that will push upper the price.Slide178

178

Random

minute and market manipulation

Now i’ll show how it could be easy to manipulate the opening price if we knew the exact ending moment of the auction. This time we need a deeper look into the auction book. Suppose, then, that what the systems have received till now is this:

If the auction ended right now, the theoretical opening price whould be 8.34 (it’s the only price that maximizes the tradable quantity, 3900 shares, see next slide).Slide179

179

Random

minute and market manipulation

Now we want to manipulate the opening price, for any reason, pushing it higher; to do that we place an order to buy 3000 shares for the auction price (at best). The book changes, and becomes like on next slide.Slide180

180

Random

minute and market manipulation

And the theoretical opening price

becomes 8.41 or 8.42. Calculating

the imbalance, then we get to the

final theoretical opening price, 8.41.Slide181

181

Random

minute and market manipulation

If we looked at the book after the auction after the manipulation (read it “what if the auction ended right now?”) we would notice that the fake order for 3000 shares with no price limits has two effects: the first is to increase the opening auction price, the second is that all the buyers offering less than 8.41 euro would be excluded from the trades. The gold rush begins!

Indeed, all the investors who were willing to buy for 8.40, or 8.37, or any other price down to 8.34 are now out of the game. And if they are really willing to buy they will have no choice but to increase the price!

Knowing that the theoretical opening price is now 8.41 they know they have to send new orders to buy for at least 8.41, better something more. Supposte, then, that all the buyers cut out from the trades by the fake order (they don’t know it’s fake, of course!) decide to modify their orders and to offer 8.43, for the same quantities requested before, 2500 shares, that are added to the 600 already there at the price 8.43.

Now suppose that one second before the end of the auction the fake order to buy 3000 shares with no price limites is cancelled. On next slide you can see what the book looks like then, and the final theoretical opening price, assuming that yesterday’s control price was 8.30 (because we have to apply the third rule to get to the opening price).Slide182

182

Random

minute and market manipulation

As we can see, the final theoretical

opening price is 8.38, not 8.34.

We were able to push up the price

without buying anything!Slide183

183

The

continuous phase

The continuous phase follows logics similar to those of the auction phase, but with some important differences. First of all, this is a very dynamic phase: any order that can be executed is executed istantaneously; if a buyer and a seller come into play offering and asking the same price they are immediately matched, for the quantity that suits them both; if the two quantities are different they will trade the common part, and one of them will have a partial execution.

In the continuous phase it is not possible to see at some time a buyer offering 10 and a seller asking 9.99, because in the continuous phase book the best buyer is always lower than the best seller, otherwise they immediately match and their orders do not appear in the book. Think, for instance, that the two proposals to buy for 10 and sell for 9.99 are sent when the best buyer in that moment offers 9.99 and the best seller asks for 10: in the exact moment the two orders are sent to the market (and they will never arrive in the exact moment) they get immediate execution, but not one with the other: the buyer hits the best seller in the book, buying for 10, the seller hits the best buyer and sells for 9.99, not necessarily in this order: it all depends on who comes first.

The logic, indeed, is still the same: “first in, first out”. And according to this mechanism it is not possible that a mistake in sending an order to the market can move the price significantly far from the previous level. Suppose, for instance, that in front of a best seller asking 10 euro per share a buyer willing to pay that amount makes a mistake and sends an order to buy for 100. That order, sent to the market, will hit the first seller in the book, since he has the right to be filled first. The price of the trade will then be 10 euro!Slide184

184

The

continuous phase

An important aspect of the continuous phase book is that when buyers and sellers agree on the price of a trade they get an immediate execution, that is, they do not appear in the book!

This because the book here shows buyers offering decreasing prices and sellers asking for increasing prices, and not a single order in the book can be filled if no one takes a step toward the opposite side. Here a typical continuous phase book:

Here the book can then be seen as a picture of the battle field, where buyers and sellers face each other waiting for their opponents to resign and go into their direction. Notice that here we also have the number of proposals: it can help in forming an idea about the average dimension of the operators; this can be a very important information, since the presence of a big operator in a book can generate pressures on the price in a specific direction (we will speak again about this when we’ll see the market manipulation in the continuous phase).Slide185

185

The

continuous phase

Let’s look back to the book on previous slide. The best buyer offers 4.55 euro for 1000 shares, the best seller asks 4.56 for 700 shares. Until new actors come into play with different proposals or one of the two opponents decide to change his mind, no trades take place.

Now imagine that after a while the seller decides to ask for a little lower price: he decides to please the best buyer; to do that he changes his order, sending it for 4.55 euro and for 700 shares. Istantaneously the system matches the two proposals: the seller sells 700 shares to the first buyer in the book. The seller is completely satisfied, the buyer just bought 700 shares: he got a partial execution. He is still in the first line of the book, willing to buy 300 shares for 4.55 euro each.

Can the buyer put a condition to his order and buy only the full amount he is willing to buy? Yes, he can, using a specific condition, which we’ll see forward.

Let’s go back to an important matter: in the continuous book we do not see the trades that take place, because in the book there are only the proposals that cannot be satisfied immediately. How can we know, then, which trades really take place? Some trading platforms show aside of their books an area called

time and sales

.

It is still a table, reporting only three fields: time, price, quantity. In it are reported all real trades, with the time of execution, the price and the number of shares traded.Slide186

186

The

continuous phase

A continuous book is full of useful information. First, if we look at the first level, the distance between the best bid and the best ask price is called

bid-ask spread

.

The spread is a measure of the liquidity of a stock; liquidity is a measure of risk, because it states the ease or the complexity to buy or sell a significant amount of shares in a specific moment, without conditioning the price’s behaviour.

An example to make things clearer. Look back to the book on slide

133

and imagine to have the need to sell quickly (for any reason) 10000 shares. There are no buyers for an amount like that, so we have two ways to sell: the first is to put a sell order for 10000 shares for 4.55 euro, selling immediately the first 1000 and waiting for the other 9000 to be filled as soon as possible (here we face two risks, the first is to have to wait for a long time, the second to be overpassed by other sellers willing to sell for lower prices, cutting us out). Or, we could hit the buyers that appear in the book, selling 1000 shares for 4.55, then 3400 for 4.54, then 1800 for 4.53 and the last 3800 for 4.52,

Doing that the weighted average price would be 4.5316 euro. Imagine to have the need to sell a million shares and the problem should be perfectly clear!Slide187

187

The

continuous phase

The liquidity of a stock is not pointed out only

by

the bid-ask spread; indeed it can be judged also from eventual jumps in the prices among different levels in the book. The following book is referred to a very thin stock, Alerion Cleanpower:

The bid-ask spread is about a 3%, very high. And between different consecutive levels in the book there are significant jumps in the prices.

Suppose to be willing to buy 5000 shares: to be sure to be filled we would need to hit the first three levels on the ask side, paying a weighted average price very high; and to sell 5000 shares it would be even worse.

Pay attention: we are talking about orders valued about 15000 euro… nothing! If to trade such a small money amount we have to burn several levels in the book, then the stock is very illiquid, that is, risky.Slide188

188

The

continuous phase

A final information that comes out from the book is the status of the pressure on one of the two sides of the market.

Look back once again to the book on slide

184:

the total amount of shares on the bid side of the book is about 45000. The total amount on the ask side is about 80000.

It seems that in this moment there is a much higher pressure on the sell side, and this means that supply is stronger than demand.

Theoretically speaking, this could be a signal that the stock is weak and is very likely going to drop; but the truth is that the book is just a picture of a specific instant in time in a trading day and the pressures on one side or the other can change significantly in a fraction of a second.Slide189

189

The

closing auction phase

At the end of the continuous phase there is another auction, wose goal is to define the closing price of the day.

The mechanism at the basis of the closing auction is the same of the opening auction, with just two differences.

First, the closing auction lasts only 5 minutes: if at the end no price is found, then the systems calculate the reference price, equal to the weighted average of the last 10% of the daily trade volume, excluding the cross orders.

Second, if to get to a closing auction price we need to use the third rule, than the reference price is not the previous days closing price, but the current day’s opening price.

Also in the closing auction there’s the random minute: from 17:30 to

17:35:59

.Slide190

190

A closer look to the book

Let’s now consider a real book to make things clear. Here there’s a ‘stamp’ of the book of the stock Fiat (source: IW Bank), shot on January the 21st 2013, 5 pm, 8 minutes and 41seconds.Slide191

191

The

book “exploded”Slide192

192

The

areas of the book in details

Zone 1 shows some information about the performance of the instrument in a specific moment, and the profit or loss of opened positions (if there are any).

On the first row, in fact, we have the current price of the instrument (Pzo), 4.7 euros in this case, the percent change in respect with the previous day’s close, +3.39% here, the quantity traded in the last trade, 5442 shares this time.

The buttons “op” and “cw” set on the right part of zone 1 are fast links to the chains of options and covered warrants written on the stock Fiat. Let’s ignore them.

On the second row there are the average charge price of the stock (Pmc), that is, the average price we paid to buy it, if applicable).

We talk about average price because if, let’s say, we purchased 1k shares at 5.0 euros each and 1k shares at 4.8 euros each, then here we would see 4.9, the weighted average of the two prices, using quantities as weights.

Then we find the total amount of shares (Q), or contracts, depending on the nature of the asset, in our portfolio (if applicable), and finally the profit or loss (PL) of the position in that moment (if applicable only, once again).Slide193

193

Zone 2 reports some useful information about the trend of the stock that day.

There we can indeed read the reference price of the previous day (Rif) and the theoretical closing auction price of the current day (ThAs), that is, the price that would be the closing price if the day was ending in that moment. We will take care of auction prices forward.

Then we can read the total traded volume till that moment in that day (Vol), that is, the total number of shares traded since the beginning of that day’s negotiations, the percent change between the theoretical closing auction price and the reference price of the previous day (Th%), the intraday maximum (Max) and the intraday minimum (Min).

The

areas of the book in detailsSlide194

194

Zone 3 is the real book.

Notice first that it is a symmetrical table: the three left columns are mirrored in respect with the three right ones.

The left side of the book is a picture of the buyers active on the market in that specific moment.

This side of the book is called

demand

, or

bid

;

denaro

in italian.

On every row we then have the total number of buyers (N), the total number of shares they all together are willing to buy (QTA), and the price – identical – anyone of them is willing to pay to buy that stock in that moment (PZO).

On the first row then we have 4 buyers, willing to buy a total amount of 7838 shares (how many for each of them is not known), all willing to pay 4.7 euros for each share.

On the second row there are 12 buyers, for a total amount of 81040 shares, offering a bit less: 4.698 euros. And so on, row after row.

The buyers are sorted top to bottom, then.

The

areas of the book in detailsSlide195

195

By doing this, on the first row there always are the best buyers from the point of view of the sellers.

Think about that: if you want to sell something on a public market you will try to find the buyer who is willing to pay more than all the others!

On the first line then we have the best buyers. The price they offer is called best bid.

Then we find the second best, the third best, and so on.

The right side of the book is the side of sellers; it is called

supply

, or

ask

;

lettera

in italian.

Here the sellers are sorted bottom to top: by doing this, on the first row there are the best sellers from the point of view of the buyers.

In other words, the sellers willing to sell at a lower price. That price is called

best ask

.

In this case the best sellers are 13, for a total amount of 30893 shares, willing to sell at 4.702 euros per share.

The

areas of the book in detailsSlide196

196

The book offers an immediate sight into the market in a specific moment, then.

In the book there are some implied informations. One is related to the pressure on the bid and on the ask side on that asset.

Summing up all the quantities on the left and on the right side, in fact, we can see whether there is more strenght on the bid (the price

could

rise) or on the ask (the price

could

drop) side.

Indeed this information is included in the book: it is synthesized in a little bar with a cursor (a small square) left free to move on it: when the pressure is higher on the right side, the cursor moves to the left, and vice-versa.

We will cover this topic again when we’ll see the iceberg orders.

Other information implied in the book refer to the minimum price variation allowed for that asset and the degree of liquidity of that asset.

Pay particular attention to the second: it’s a clear measure of risk, since the higher the liquidity of the asset we need to buy or sell, the best execution we will be able to get.

The

areas of the book in detailsSlide197

197

Now a question could arise: if the best buyer is willing to pay 4.7 euros and the best seller is asking for 4.702, how can ever take place a trade?

That’s a logical question: until either the buyers or the sellers decide to settle for a little worse price, a little higher for the firsts, a little lower for the seconds, there will not be any trade!

The only way to have a trade is that someone decides to move towards his opponent.

How can we know anything about the real trades, then? What about the prices of those trades? What about the traded quantities?

These information come with zne 4, called

times and sales

.

Here indeed we find the details of all the real trades that take place, sorted bottom to top (on the top of the table we can see the most recent trades); when a new trade takes place, it shows up on the first line of the times and sales, pushing down all the previous trades.

The

areas of the book in detailsSlide198

198

Referring to the example, 5442 shares (notice that it’s the same value that shows up in zone 1, since it’s the last trade) have been traded at 5 hours, 8 minutes, 41 seconds, for 4.7 euros per share.

Eleven seconds earlier 4072 shares were traded, for the same price.

Two more seconds earlier other 872 shares, again for the same price. And so on.

The times and sales is fundamental, because it tells us something about the real pressures acting on an asset at some specific moment, and in which direction.

This concept will become clearer in a while, when we’ll talk about the iceberg orders.

Another natural question: who sells (or buys) first?

The book is managed by a computer that collects all the proposals and aggregate them on the same price levels, respecting the crhonological order of appearance.

The general rule is that who comes first has the right to be executed first: first in, first out.

The

areas of the book in detailsSlide199

199

There’s an exception to the rule “first in, first out”, though.

This exception shows up when operators make proposals given of specific limits in terms of quantity (we will take care of this issue later).

A problem that can show up in the trading activity is that the quantity asked or offered get satisfied only partially with a single trade.

Such issues are referred to as partial executions.

Think, for example, to be the first of the buyers in the best bid position, and to be trying to buy 1000 shares.

If a seller decides to settle for our price and hits us, but he is willing to sell only 750 shares he has the right to be executed, because he comes first. He will sell 750 shares, then, and we will remain on the first line of the bid side of the book with the residual 250 shares, until a second seller shows up.

What if no one else is willing to sell at our price (that is, the price doesn’t go down any further)? We may remain with a partial execution.

The

areas of the book in detailsSlide200

200

Zone number 5 is where we can place orders to buy or sell.

On the first row there comes the simbol of the asset (named

ticker

): each single asset has its own label, so that it can be identified without any doubt. Then we have the field of the quantity we are willing to buy or sell, the price asked, and the type of the price.

The type is set by default on the option

price limit

(Limite Pzo). We will take care of this parameter very soon.

Right under we have button for the cancellation of all the

pending

(revoca), that is, not yet executed, orders (if there are any), the button for buying (acquista) and the button for selling (vendi). Finally we have the button

chiudi posizione al meglio

, CaM, that closes immediately all the opened positions (if there are any) and the button

chiudi posizione a limite di prezzo,

CaP, that sends to the system an order to close all the opened positions (if there are any) at a specific price. We will take care of this issue too, very soon.

The simbol “+” set on the bottom right of the book opens more options. Let’s move on for now.

It’s time to take a step into the magic world of operative orders!

The

areas of the book in detailsSlide201

201

Order

types

Orders can be of several types:

at market

(or

best

),

limit price

,

conditional price

(also called

stop orders

).

They can also be enriched with parameters about the quantity:

all or nothing

,

visible quantity

.

And given of a specific time validity as well:

good till cancel

, or

valid till date

.

The most common order type is the one so called

at best

.

To buy or sell at best means to hit the best seller or the best buyer that appear in the book in that specific moment. It is said to buy or sell at best because it means to hit the best bid or the best ask.

To do that we just need to write down the number of shares in the proper field, leave empty the filed of the price, and then click on the button buy or sell.

We will get an instant execution (it really takes just a few milliseconds), that will be higlighted by a sound of some sort in the trading platform. Our order will not even for a second appear in the book: it will immediately appear in the times and sales.

Immediately after, in zone 1 of the book we will see: the number of shares purchased, the price paid, and the real time performance of the position.Slide202

202

Step 8: Market orders

To be fussy, actually, in the definition “at best” there is a contraddiction in terms.

An exchange is indeed some sort of fish market, where buyers and sellers meet and fight to get the best price ever for them.

To buy or sell at best means to give up to the opportunity to try and get something better.

But to be honest, if no one ever decided to settle for something a bit worse, there would not be any trade: the market would not even exist!

The advantage of a market order is that we can have the immediate confirmation about the execution of that order.

Looking back at the case of Fiat, to buy at best would mean to buy shares at 4.702 euros each; to sell at best would mean to sell shares at 4.7 euros each.Slide203

203

Step 8: Price limit orders

An alternative consists in the choice of a so called price limit order.

Suppose for example to be willing to buy 1k Fiat shares at 4.692 euors each instead of 4.702. In order to do that we have to write 4.692 in the price field before clicking on the buy button.

Once we have clicked on the order confirmation we will appear on the left side of the book, fifth row (assuming that the book is still the one seen previously, of course): the number of buyers will be increased by 1 and the quantity will be increased by 1k.

There are an advantage and a disadvantage in this choice. The advantage is that if our order is executed we will buy at a price lower than the one asked by the best sellers in this moment. The disadvantage is that there is no certainty to be executed: if the price doesn’t decrease down to 4.692 we will never buy!

Until we get executed the order is pending. If an order remains pending till the markets close it is automatically cancelled, unless in the moment it was sent to the market it was given of a different specific in terms of time validity (see next slide). Slide204

204

Step 8: Time validity of orders

Apart from specific indications, all the orders placed on the market in some day remain valid till the market close; then, if they have not been executed yet, they are automatically canceled.

Such orders are called

good till cancel

(GTC), because as a matter of fact who place them on the market can also decide to cancel them at any time, without the need to wait for the market close.

If the idea is to keep an order valid after the market close, for one or more days, we can use an order

valid till date

: before sending the order to the market we can tell the system the last date of validity writing it down in a specific field.

At the market close, every day, the system automatically cancels the orders not yet executed, and the next day, as the market opens, the system automatically sends again the order to the market. Slide205

205

Step 8: Conditional orders

A conditional order, often referred to as

stop order

, can be used for several purposes.

The result is to exponentially increase the number of options available for the trader.

Let’s see an example to make things clear.

Suppose a stock has an important resistance at the price 10.72 euros. If that important price level gets crossed by an upward move it is very likely that the strenght of the move will increase, since the breakout will probably attract many investors.

Until that threshold keeps pushing down the price, there are no interesting signals, hence we need to monitor the price waiting for that level to be crossed, ready to buy when this event shows up.

But this means we have to stay hours or days in front of the sceen waiting for the breakout. We need to find another way.

The solution comes from stop orders (see next slide).Slide206

206

Step 8: Conditional orders

What we want to put in place now is an order of this sort: “if last price is higher than 10.72 then send to the market an order to buy 1k shares at best”.

In order to do that we need to write 1000 in the quantity field, leave empty the price field, set, if we want to, a time validity for the order (the condition to buy will not probably be satisfied today), and finally set the condition that, if satisfied, makes the order valid. The condition is made of two components: the type of the condition and the trigger (the conditioned price).

The type of condition in this case is “last price > than”, or, maybe “last price >= than” (it depends on the trading platform).

The trigger, 10.72 in this case (10.73 if the condition is >=, since it’s on the breakout of the price 10.72 that we want to buy, not at that same level), has to be set in the proper filed.Slide207

207

Step 8: Multi-conditional orders

In the moment we click on the “buy” button and we confirm the order, the order will not be sent to the market: it will remain hidden on the bank’s servers, waiting for the trigger to be hit.

If we gave the order a time validity, the order will remain inactive, but ready to go, until either the trigger or the expiration of the order is reached.

If the trigger has never been reached before the maturity of the order, at that time the order will automatically be canceled, together with the condition.

An extremely interesting variant of conditional orders is given by the opportunity to set not just an order to buy conditioned to the occurrence of a specific condition, but also two more orders, dependent from the first, for the automatic stop loss, if things go bad, or take profit, if things go well.

Let’s make an example. Suppose then to be willing to buy shares of the stock XYZ if the price crosses the 10.72 resistance within 10 days, and that in case of execution we would like to take profit if the price reaches 11.50, or to cut losses if the price drops under 10.45.

This is a complete strategy: entry price with a specific time validity, stop loss and take profit. Entry strategy, timeline, risk management, profit management, all in one.Slide208

208

Step 8: A full trade via book

In order to do that we just need to write 10.45 in the field “stop loss”, and 11.50 in the field “take profit”.

Once orders have been confirmed one single order (the so called

father order

) will be ready to get executed: the entry order. The bank’s systems will start monitoring the trigger price; in the moment it gets crossed the market order to buy will be automatically sent to the market.

Until this happens the other two orders (called

sons orders

) are hidden and inactive.

In the moment the father order is executed the system automatically starts monitoring the two new triggers: the stop loss and the take profit.

IN the moment one of them is reached the system automatically sends the correspondent market order, and cancels the other one.

The two sons orders are indeed self-excludent: in the moment one of them is executed, the other one is automatically canceled. They are called OCO (one cancels other).Slide209

209

Step 8: All or nothing orders (tutto o niente in italian)

Previously we stated that the first in first out rule can have some exceptions.

One of these exceptions is given by the all or nothing orders (ToN), with which we can tell the market that we will not accept to be satisfied only partially on our order: if, for example, we are willing to buy 10k shares and the best seller offers only 2k shares he will not sell his shares to us but to the buyer immediately after us.

In this way the best seller will apply his right to be executed for first; on the other hand we will slip in second position until we will find a seller that matches at least the number of shares we are willing to buy.

Our order is always ideally in the first place in terms od priority, but it will be executed only whwn it could be fully executed in one single shot.

Until that happens we will be climbed over by the second best buyers.Slide210

210

Step 8: Iceberg orders

Let’s look back at the book of Fiat and make some quick sums. The total amount asked by all the buyers is 211833 shares. The total amount offered by the sellers is 301640 shares.

Now think you are the manager of a huge investment fund, willing to buy 1 million Fiat shares, for any reason.

Imagine to place a limit price order for 1M shares, let’s say for 4.694 euros each. What you think will happen?

A clue: the book is the expression of the balance between supply and demand acting on the market in a specific moment: at any time in front of the book there may be millions of investors set in every part of the world.

To figure out what it would happen we have to identify ourselves with any investor who is looking at the book in that moment: he can see few thousands of shares traded per trade (that’s what the times and sales tells him), but suddenly… a single buyer for 1M shares…Slide211

211

Step 8: Iceberg orders

The most obvious consequence is that the price goes up like a rocket, because the whole market understands that there’s a huge investor interested in that stock: usually no one invests millions of euros without a good reason…

We can be pretty sure that the stock is going to go up sharply.

The result? The price will rais so fastly that who wanted to buy 1M shares will not get any!

This issue can be solved in two ways. The first consists in placing manually several smaller orders, let’s say 10k shares a time, one at a time: when the first is executer we place a second one, then a third, and so on, until we get executed for 1M shares.

The alternative, much better, is to tell the system to do that on its own.

In order to do that we need to take advantage of the function QtaVis (visible quantity), that can be activeated by an appropriate field in the book.

The fields to be filled are two: the total number of shares to buy (or sell) and the visible quantity.Slide212

212

Step 8: Iceberg orders

Once we have clicked on the confirm button, the system will send to the market an order to buy or sell 10k shares; once executed it will send a second order, and so on, till the total amount of shares has been bought (or sold).

Such orders are called iceberg (asteriscati in italian), because what appears in the book is just the summit of a much bigger order, hidden, just to say, under the surface of the sea.

Of course after a while the market can realize that there is an iceberg order, since every time 10k shares are traded at 4.694 euros each a new order for 10k shares shows up.

The problem (from the point of view of other operators) is that even if they realize there is an iceberg order they cannot know what is the total amount underneath it, that is, they cannot form an idea about th size of the buyer. Slide213

213

Step 8: stock manipulation in the continuous phase

The book can be used for illicit purposes too.

In order to explain this issue let’s make an example.

Suppose again to be the managers of an investment fund, willing to buy 1M Fiat shares (the same applies on the other side, mutatis mutandis). We do not want the price to go up like a rocket before we buy.

A way to do that is using an iceberg order, of course. But there could be some problems: iceberg orders not always are fulfilled, since after a while the market realizes that something is going on and no one buys or sells anymore before they understand what’s happening.

There is also another way, forbidden by the law, but nonetheless very popular: the market manipulation.Slide214

214

If i wanna have a chance to buy 1M shares without pushing up the price i need to make the market believe there is someone else willing to sell an equal – or even higher – amount on the other side.

This is how it works: we place an order to sell 1M – or more – shares visible on the book, but in a safe position, let’s say row 4 or 5.

The idea is that in the moment the sell order shows up in the book the price will drop, because all the operators will sell at best just to be sure to get out and leave before the others.

We will not sell a share, so, but we do not want to!

Our real purpose is to push down the price, so to have a reasonable certainty that we will be able to buy a huge amount of shares at tre price we decide to buy.

It is,

remember,

a forbidden behaviour. Nonetheless it always happens, probably

because (some say)

it is quite hard to prove the intention to manipulate the price.

Step 8: stock manipulation in the continuous phase