Event Studies 1 An event study is designed to examine market reactions to and abnormal returns around specific informationimparting events These events can be marketwide or firmspecific The event can occur at the same point in time for all stocks 911 or at different points in time for each ID: 765773
Download Presentation The PPT/PDF document "Event Studies 1 An event study is design..." 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.
Event Studies 1
An event study is designed to examine market reactions to, and abnormal returns around specific information-imparting events. These events can be market-wide or firm-specificThe event can occur at the same point in time for all stocks (9/11) or at different points in time for each stock (M&A Announcement) Event Study 2
The day or time of the event is day (or time) 0. You are looking to see if there is unusual movement either on day 0, before day 0, or after day 0. You want to find out if there is convincing evidence that the event caused the unusual movement. Knowing that an event caused or didn’t cause an abnormal return can help us predict stock reactions to similar events in the future. Event Study 3
1. Clearly identify the event and the date on which the event announcement occurred Note that the announcement date is preferred to the date that the event took place if you expect prices to react to the announcement 2. Find returns around the event date Usually daily, but can be more frequent (CNBC Reports) or less frequent 3. Determine your sample size and collect all the pertinent data in your sample How to do an Event Study 4
4. Determine the event window Days –n to +n 5. Control for the Market Determine how much of each stock’s movement was due to market movement separate from the event. This will give us the excess returns we are looking for. 6. Average the excess returns across all stocks in the sample for each day in the event window Equally-weighted or value-weighted How to do an Event Study 5
7. Calculate the standard errors of the excess returns Sample standard deviation divided by the square root of the sample size 7. Calculate t-statistics to test for statistical significance Divide the average abnormal return for the day by the standard error and compare with traditional hurdle rate of t=1.96 8. Plot cumulative abnormal returns on a graph to give a clear visual description of what you found. How to do an Event Study 6
9. If desired, calculate a long-short portfolio Determine the results of a portfolio where you go long the top decile (or 20%) and short the bottom decile (or 20%). This works best if the top decile has positive abnormal returns and the bottom decile has negative abnormal returns How to do an Event Study 7
No Risk Adjustment Actual Rit – Actual R m,t Often close enough for short-term movements Risk Adjustment using Alpha and Beta Actual R it – [ai + Bi [Actual R m,t ] a i and Bi must be determined outside event window Risk Adjustment using Matched Firms Actual R it – [Actual Rmatch,t]Find matches in same industry with closest possible market cap 8 Controlling for Market Movement
M & A AnnouncementsHow do stocks react before, at, and after an announcement that there is a bid to acquire them? IPO Long-Run ReturnsHow do stocks do over the three years following their IPO? Do those with higher initial returns do better than those with lower initial returns? Reaction to CNBC Reports If a stock is mentioned favorably or unfavorably on CNBC, does it affect its price? Is so, how quickly? 9 Examples of Event Studies
Momentum in Weekly Returns If we long the stocks with the top decile of weekly returns and short the bottom decile, how will our portfolio perform over the next 52 weeks? Stocks Within Mutual Funds Ranking mutual funds based on their inflow of new money, we can form a portfolio where we long the stocks that the top 20% hold and short the stocks that the bottom 20% hold. How well does this portfolio do over the next 40 months? 10 Examples of Event Studies
Name ChangesWas there an abnormal return when companies changed their name to include “dot com” between January 1998 and March 1999? 11 Examples of Event Studies
M&A Announcements 12
13 IPO Long-Run Returns
Reaction to CNBC Reports 14
Short-Term Momentum 15
The Dumb Money 16
17 Name Change
Perform an Event StudyFind out if being added to the S&P 500 has an impact on a stock’s returnRandomly select 100 stocks that were added to the S&P 500 between 1/2000 and 8 /2015 Stocks and dates are provided Price data comes from Yahoo Finance Market model is used to find abnormal returns CARs for Event Day-10 to Event Day+10 CAVs (Cumulative Abnormal Trading Volume) 18 Assignment