/
THREE NOVEL FACTORS AFFECTING HEDGE FUND RETURNS China Derivatives Market Conference (CDMC) THREE NOVEL FACTORS AFFECTING HEDGE FUND RETURNS China Derivatives Market Conference (CDMC)

THREE NOVEL FACTORS AFFECTING HEDGE FUND RETURNS China Derivatives Market Conference (CDMC) - PowerPoint Presentation

lois-ondreau
lois-ondreau . @lois-ondreau
Follow
345 views
Uploaded On 2019-11-01

THREE NOVEL FACTORS AFFECTING HEDGE FUND RETURNS China Derivatives Market Conference (CDMC) - PPT Presentation

THREE NOVEL FACTORS AFFECTING HEDGE FUND RETURNS China Derivatives Market Conference CDMC May 20 2016 Suzhou China L Mick Swartz PhD CAIA University of Southern California Marshall School of Business ID: 762069

model variable coefficient opt variable model opt coefficient intercept criterion info ols ratio amp p500 factor return sml estimation

Share:

Link:

Embed:

Download Presentation from below link

Download Presentation The PPT/PDF document "THREE NOVEL FACTORS AFFECTING HEDGE FUND..." 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

THREE NOVEL FACTORS AFFECTING HEDGE FUND RETURNSChina Derivatives Market Conference (CDMC)May 20, 2016 – Suzhou, China L. Mick Swartz (PhD, CAIA) University of Southern California Marshall School of Business

Presentation OutlineConceptual Introduction to Three New Ratios Drawdown and Run-up Variables Graphical IllustrationL-Ratio as a Liquidity Factor R-Ratio as a Momentum Factor D-Ratio as an Average Downside Risk Factor Fung & Hsieh Seven-Factor Model Data Collection, Model Estimation & Diagnostics

Presentation Outline – continueEmpirical ResultsGlobal Macro StrategiesEmerging Market Strategies Asia Regional Strategies Equity Hedge Strategies Relative Value Strategies Event Driven Strategies Europe Regional Strategies Final Remarks

Conceptual Introduction to Three New RatiosThe three new ratios are generated with hedge fund strategy’s monthly excess return in the numerator, as is the case for traditional asset allocation ratios such as Sharpe ratio. The denominator of these ratios includes risk factors that are considered to be more relevant and vital for hedge fund industry, such as Drawdown, Run-up, and their corresponding magnitude, duration, and velocity, as opposed to Std. Dev., Beta etc. L-Ratio is used as a Liquidity factor, i.e. how long (no. of months) it takes the return of hedge fund to recover from it’s highest fall as the risk factor R-Ratio is used as a Momentum factor, i.e. how fast the hedge fund return rises from it’s last drop as the risk factor D-Ratio is used as an Average Drawdown% (Downside Risk) factor, i.e. the average drawdown% per month as the risk factor

Drawdown & Run-up Variables Graphical Illustration

L-Ratio as a Liquidity FactorL-Ratio as a Liquidity Factor  

R-Ratio as a Momentum FactorR-Ratio as a Momentum Factor  

D-Ratio as an Average Downside Risk FactorD-Ratio as an Average Downside Risk Factor  

Fung & Hsieh Seven-Factor ModelFung & Hsieh 7-Factor Model Variables*:S&P500: Standard & Poor's 500 stock return. SML: Russell 2000 Small Cap return  Russell 1000 Large Cap return T10Y: Month end-to-month end change in the Federal Reserve’s 10 year constant maturity yield. Credit Spread: month end-to-month end change in the difference between Moody’s Baa yield and the Federal Reserve’s 10 year constant maturity yield.Bd. Opt: return of a portfolio of lookback straddles on bond futures.FX Opt: return of a portfolio of lookback straddles on currency futures.Com. Opt: return of a portfolio of lookback straddles on commodity futures.Note: if 5,6 or 7 are significant, then GARCH should be implemented. * Fung, W., Hsieh, D., 2004. Hedge fund benchmarks: a risk-based approach. Financial Analyst Journal 60(5), 65–80.

Data Collection, Model Estimation & DiagnosticsData Collection and Variable ConstructionMonthly return of 55 HFRX hedge fund strategies collected from “HFR Inc.”Monthly total return of S&P500 index, CRB index, MSCI World index, Russell 2000 index and Russell 1000 growth and value indices (for SML and HML), and other countries’ market indices, are collected from “ Global Financial Data ”. 3-month and 12-month LIBOR rates, U.S. 10-year T-Bond and 3-month T-Bill, and Moody’s Baa Corp. Bond yield for Credit Spread, are collected from “ Federal Reserve Bank of St. Louis”. Statistical variables such as Std. Dev., Skew, excess Kurtosis, Sharpe Ratio, Sortino Ratio, Max. Drawdown percentage, Duration, and Velocity; Max. Run-up percentage, Duration, and Velocity, and new ratios are calculated and used as independent variables.

Data Collection, Model Estimation & DiagnosticsOther New Ratios Note: S 2  L-Ratio , S 6  R-Ratio , S 7  D-Ratio  

Data Collection, Model Estimation & DiagnosticsModel Estimation and DiagnosticsHedge fund return models are estimated by Ordinary Least Square (OLS) technique or one of the ARCH family techniques if needed.To generate robust models, each model is tested and corrected for: Stationarity (Augmented Dickey-Fuller Unit Root test, corrected by first differencing), Serial-Correlation (Durbin-Watson test, corrected by first order a utoregression term), Multi-Collinearity (Variance Inflation Factor test < 10, corrected by elimination ), Conditional Heteroskedasticity (tested and corrected by ARCH, GARCH, or EGARCH modeling). Heteroskedasticity (White test, corrected by Newey–West HAC estimation).The Akaike information criterion (AIC) and Schwarz information criterion (SIC) are used for the best model selection, with the SIC favored, as an indicator of the parsimony model, if there is a disagreement among these indicators.

Global Macro StrategiesEmpirical Results

Macro/CTADependent Variable: Macro/CTA return   Sample Period: January 1998 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.003 Intercept − 0.002 S&P500 0.01 L-Ratio − 0.80* SML 0.10** S 1 0.02* T10Y − 0.53 Credit Spread 0.33 Bd. Opt 2.10 FX Opt − 0.03 Com. Opt 0.11* Adj. R-squared 8.15% 24.25% Akaike info criterion (AIC) − 4.68 − 5.17 Schwarz info criterion (SIC) − 4.55 − 5.06 Estimation Technique OLS EGARCH Significance Levels: *** 10% and ** 5% and * 1%

Macro: Active TradingDependent Variable: Macro: Active Trading return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.005** Intercept −0.008* S&P500 − 0.002 D-Ratio 0.001* SML − 0.10*** T10Y 9.82 Credit Spread 0.08 Bd. Opt 0.01 FX Opt − 0.03 Com. Opt 0.06*** Adj. R-squared 5.82% 20.90% Akaike info criterion (AIC) − 5.81 − 6.13 Schwarz info criterion (SIC) − 5.62 − 6.03 Estimation Technique OLS GARCH Significance Levels: *** 10% and ** 5% and * 1%

Macro: CommodityDependent Variable: Macro: Commodity return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.003 Intercept − 0.084* S&P500 − 0.09 R-Ratio 6.98* SML 0.09 CRB 0.12** T10Y 12.90 Max Drawdown% − 0.33* Credit Spread 0.60 S 5 0.76* Bd. Opt 0.03 D-Ratio 0.002* FX Opt − 0.03 Com. Opt 0.17* Adj. R-squared 20.87% 40.28% Akaike info criterion (AIC) − 5.12 − 5.43 Schwarz info criterion (SIC) − 4.90 − 5.30 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Macro: Commodity–AgricultureDependent Variable: Macro: Commodity–Agriculture return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.003 Intercept − 0.016* S&P500 − 0.02 R-Ratio 2.81* SML − 0.01 CRB 0.12* T10Y − 3.33 Kurtosis 0.006* Credit Spread − 0.14 Bd. Opt 0.10 FX Opt − 0.02 Com. Opt 0.15* Adj. R-squared 6.37% 24.10% Akaike info criterion (AIC) − 4.75 − 5.05 Schwarz info criterion (SIC) − 4.56 − 4.91 Estimation Technique OLS ARCH Significance Levels: *** 10% and ** 5% and * 1%

Macro: Commodity–EnergyDependent Variable: Macro: Commodity–Energy return   Sample Period: January 2007 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept 0.002 S&P500 − 0.04 CRB 0.24* SML − 0.14 D-Ratio 0.003* T10Y 23.80 L-Ratio − 1.23* Credit Spread 0.41 Bd. Opt 2.49 FX Opt 0.17 Com. Opt 0.40* Adj. R-squared 27.68% 47.08% Akaike info criterion (AIC) − 4.12 − 4.47 Schwarz info criterion (SIC) − 3.90 − 4.36 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Macro: Commodity–MetalsDependent Variable: Macro: Commodity–Metals return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.005 Intercept − 0.003 S&P500 − 0.23 CRB 0.63* SML 0.48** R-Ratio 7.97* T10Y − 15.30 Run-up Duration − 0.001* Credit Spread 7.26 Drawdown Duration 0.002* Bd. Opt 6.72 FX Opt − 0.16 Com. Opt 0.78* Adj. R-squared 34.12% 37.17% Akaike info criterion (AIC) − 2.88 − 2.95 Schwarz info criterion (SIC) − 2.69 − 2.83 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Macro: CurrencyDependent Variable: Macro: Currency return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.001 Intercept 0.04* S&P500 0.04 Sortino Ratio 0.08* SML 0.02 D-Ratio 0.001* T10Y 15.53 S 4 − 0.05* Credit Spread − 0.13 Bd. Opt 0.28 FX Opt − 0.04 Com. Opt − 0.02 Adj. R-squared 0.80% 22.30% Akaike info criterion (AIC) − 5.76 − 6.04 Schwarz info criterion (SIC) − 5.57 − 5.94 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Macro: Discretionary ThematicDependent Variable: Macro: Discretionary Thematic return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.004 Intercept − 0.046* S&P500 0.03 CRB 0.22* SML 0.02 Max Drawdown% − 0.15* T10Y − 26.71** R-Ratio 8.17* Credit Spread 0.89 L-Ratio − 0.47* Bd. Opt − 1.11 S 5 0.84* FX Opt 0.07 Com. Opt 0.30* Adj. R-squared 36.88% 51.47% Akaike info criterion (AIC) − 5.00 − 5.28 Schwarz info criterion (SIC) − 4.81 − 5.13 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Macro: Multi-StrategyDependent Variable: Macro: Multi-Strategy return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.0002 Intercept − 0.01*** S&P500 0.11* S&P500 0.16* SML − 0.01 Run-up Velocity 0.001* T10Y 1.03 Credit Spread 3.15** Bd. Opt − 0.68 FX Opt 0.01 Com. Opt 0.08** Adj. R-squared 19.46% 17.94% Akaike info criterion (AIC) − 5.54 − 5.56 Schwarz info criterion (SIC) − 5.35 − 5.49 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Macro: Systematic DiversifiedDependent Variable: Macro: Systematic Diversified return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.005 Intercept 0.03** S&P500 − 0.20** Sortino Ratio 0.05* SML 0.02 Run-up Velocity − 0.003* T10Y 18.15 L-Ratio − 0.98* Credit Spread 0.02 S&P500 − 0.14** Bd. Opt 3.94** FX Opt 0.04 Com. Opt 0.12*** Adj. R-squared 7.00% 19.11% Akaike info criterion (AIC) − 4.30 − 4.47 Schwarz info criterion (SIC) − 4.12 − 4.35 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Emerging Market StrategiesEmpirical Results

Emerging Markets Composite IndexDependent Variable: Emerging Markets Composite Index return   Sample Period: January 2006 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.003 Intercept − 0.04* Intercept − 0.05* MSCI 0.33* MSCI 0.23* MSCI 0.33* SML − 0.13 R-Ratio 6.06* HML (Value Prem.) − 0.35* T10Y − 0.88 Max Drawdown% − 0.10* R-Ratio 6.02* Credit Spread 3.60** Max Drawdown% − 0.11* Bd. Opt 0.58 FX Opt 0.18 Com. Opt 0.18* Adj. R-squared 52.61% 49.46% 58.76% Akaike info criterion (AIC) − 5.11 − 5.31 − 5.28 Schwarz info criterion (SIC) − 4.91 − 5.13 − 5.15 Estimation Technique OLS GARCH OLS Significance Levels: *** 10% and ** 5% and * 1%

Total Emerging Market IndexDependent Variable: Total Emerging Market Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept − 0.02* Intercept − 0.03* MSCI 0.36* MSCI 0.45* MSCI 0.44* SML − 0.09 Sharpe Ratio 0.01* HML (Value Prem.) − 0.25* T10Y 6.45 Kurtosis 0.004* Sharpe Ratio 0.02* Credit Spread 0.88 Kurtosis 0.006* Bd. Opt − 0.98 FX Opt − 0.03 Com. Opt 0.13* Adj. R-squared 75.83% 75.19% 78.29% Akaike info criterion (AIC) − 5.77 − 5.92 − 5.90 Schwarz info criterion (SIC) − 5.58 − 5.78 − 5.78 Estimation Technique OLS ARCH OLS Significance Levels: *** 10% and ** 5% and * 1%

Multi-Emerging Markets IndexDependent Variable: Multi-Emerging Markets Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept − 0.001 Intercept − 0.02* MSCI 0.51* MSCI 0.43* MSCI 0.51* SML − 0.19 D-Ratio 0.001** HML (Value Prem.) − 0.39* T10Y − 19.60 Sortino Ratio 0.02* Credit Spread 1.00 Max Drawdown% − 0.09* Bd. Opt − 1.20 Max Run-up% − 0.03* FX Opt 0.20 Com. Opt 0.18* Adj. R-squared 71.89% 66.67% 75.99% Akaike info criterion (AIC) − 5.23 − 5.32 − 5.40 Schwarz info criterion (SIC) − 5.04 − 5.18 − 5.26 Estimation Technique OLS GARCH OLS Significance Levels: *** 10% and ** 5% and * 1%

Asia ex-Japan IndexDependent Variable: Asia ex-Japan Index return   Sample Period: January 2004 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.001 Intercept − 0.001 Intercept − 0.005 MSCI 0.32* MSCI 0.46* MSCI 0.46* SML 0.04 S 3 3.26* S 3 4.18* T10Y − 2.90 AR(1) 0.30* HML (Value Prem.) − 0.26* Credit Spread 2.85 AR(1) 0.29* Bd. Opt − 1.24 FX Opt − 0.03 Com. Opt 0.19* Adj. R-squared 42.69% 55.16% 56.19% Akaike info criterion (AIC) − 4.48 − 4.90 − 4.89 Schwarz info criterion (SIC) − 4.28 − 4.72 − 4.71 Estimation Technique OLS EGARCH GARCH Significance Levels: *** 10% and ** 5% and * 1%

Brazil IndexDependent Variable: Brazil Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.004 Intercept 0.0004 BOVESPA 0.40* BOVESPA 0.48* SML 0.07 T10Y 15.25 Credit Spread − 1.80 Bd. Opt 0.46 FX Opt − 0.11 Com. Opt 0.11*** Adj. R-squared 63.60% 61.51% Akaike info criterion (AIC) − 4.57 − 4.56 Schwarz info criterion (SIC) − 4.38 − 4.52 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

BRIC IndexDependent Variable: BRIC Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.001 Intercept − 0.15* Intercept − 0.15* MSCI 0.71* MSCI 0.68* MSCI 0.69* SML − 0.11 S 1 0.02* S 1 0.02* T10Y 0.82 Std. Dev. 3.04* HML (Value Prem.) − 0.33* Credit Spread 3.20 L-Ratio − 0.45* Std. Dev. 2.96* Bd. Opt − 0.57 S 4 0.03* L-Ratio − 0.49* FX Opt 0.22 S 4 0.03* Com. Opt 0.17* Adj. R-squared 68.82% 79.58% 81.53% Akaike info criterion (AIC) − 4.48 − 4.98 − 5.03 Schwarz info criterion (SIC) − 4.27 − 4.74 − 4.86 Estimation Technique OLS EGARCH OLS Significance Levels: *** 10% and ** 5% and * 1%

China IndexDependent Variable: China Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept 0.002 Intercept 0.001 Shanghai Comp 0.18* Shanghai Comp 0.14* Shanghai Comp 0.13* SML 0.17 D-Ratio 0.01* D-Ratio 0.01* T10Y 1.98 Sharpe Ratio 0.04* R-Ratio 8.19* Credit Spread 4.97** R-Ratio 8.93* Sharpe Ratio 0.03* Bd. Opt − 1.83 HML (Value Prem.) − 0.29* FX Opt − 0.15 SML (Size Prem.) 0.19** Com. Opt 0.16** Adj. R-squared 44.71% 57.62% 60.36% Akaike info criterion (AIC) − 4.37 − 4.66 − 4.71 Schwarz info criterion (SIC) − 4.18 − 4.54 − 4.54 Estimation Technique OLS OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

India IndexDependent Variable: India Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.005 Intercept − 0.0003 Nifty 0.79* Nifty 0.80* SML − 0.27** D-Ratio 0.009* T10Y 37.97 Credit Spread − 2.71 Bd. Opt − 1.03 FX Opt − 0.56* Com. Opt − 0.12 Adj. R-squared 78.41% 77.16% Akaike info criterion (AIC) − 4.03 − 4.01 Schwarz info criterion (SIC) − 3.84 − 3.94 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Korea Index Dependent Variable: Korea Index return   Sample Period: January 2008 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.006 Intercept − 0.0004 KOSPI 0.54* KOSPI 0.67* SML 0.20 T10Y 22.20 Credit Spread − 3.56 Bd. Opt 5.63* FX Opt − 0.58* Com. Opt 0.002 Adj. R-squared 73.75% 62.47% Akaike info criterion (AIC) − 4.46 − 4.16 Schwarz info criterion (SIC) − 4.23 − 4.10 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Latin America IndexDependent Variable: Latin America Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.0001 Intercept − 0.004 MSCI 0.40* MSCI 0.47* SML − 0.08 D-Ratio 0.002* T10Y − 0.31 AR(1) 0.26* Credit Spread 2.38 Bd. Opt − 1.44 FX Opt 0.05 Com. Opt 0.16* Adj. R-squared 63.46% 61.67% Akaike info criterion (AIC) − 5.07 − 5.06 Schwarz info criterion (SIC) − 4.85 − 4.96 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

MENA (Middle East/North Africa) Index Dependent Variable: MENA Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.003 Intercept − 0.001 MSCI 0.50* MSCI 0.55* SML 0.03 D-Ratio 0.001* T10Y 28.02** Credit Spread 1.79 Bd. Opt − 1.21 FX Opt 0.13 Com. Opt 0.11* Adj. R-squared 76.04% 71.35% Akaike info criterion (AIC) − 5.36 − 5.22 Schwarz info criterion (SIC) − 5.17 − 5.15 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Russia IndexDependent Variable: Russia Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.005 Intercept − 0.007*** Intercept 0.002 MICEX 0.51* MICEX 0.57* MICEX 0.57* SML − 0.02 Sharpe Ratio 0.03* HML (Value Prem.) 0.29** T10Y 23.35 R-Ratio 4.56* D-Ratio 0.01* Credit Spread − 1.96 AR(1) 0.38* Bd. Opt − 0.05 FX Opt − 0.30* Com. Opt 0.09 Adj. R-squared 79.88% 78.91% 79.10% Akaike info criterion (AIC) − 4.37 − 4.50 − 4.49 Schwarz info criterion (SIC) − 4.16 − 4.31 − 4.30 Estimation Technique OLS GARCH GARCH Significance Levels: *** 10% and ** 5% and * 1%

Russia/Eastern Europe IndexDependent Variable: Russia/Eastern Europe Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.007 Intercept − 0.03* MSCI 0.52* MSCI 0.61* SML − 0.08 R-Ratio 9.69* T10Y 20.08 L-Ratio − 0.58* Credit Spread − 4.13 Bd. Opt − 1.09 FX Opt − 0.01 Com. Opt 0.14*** Adj. R-squared 58.37% 67.10% Akaike info criterion (AIC) − 4.23 − 4.51 Schwarz info criterion (SIC) − 4.02 − 4.42 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Asia Regional StrategiesEmpirical Results

Asia Composite Hedge Fund IndexDependent Variable: Asia Composite Hedge Fund Index return   Sample Period: January 2004 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.0003 Intercept − 0.003 Intercept − 0.005** MSCI 0.37* MSCI 0.36* MSCI 0.34* SML − 0.03 R-Ratio 1.44** R-Ratio 1.91* T10Y 14.04 HML (Value Prem.) − 0.15** Credit Spread 2.06 Bd. Opt 0.12 FX Opt 0.16** Com. Opt 0.06 Adj. R-squared 52.57% 52.05% 54.50% Akaike info criterion (AIC) − 5.39 − 5.57 − 5.55 Schwarz info criterion (SIC) − 5.19 − 5.41 − 5.40 Estimation Technique OLS EGARCH GARCH Significance Levels: *** 10% and ** 5% and * 1%

Asia Equally Weighted IndexDependent Variable: Asia Equally Weighted Index return   Sample Period: January 2004 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.001 Intercept − 0.005*** MSCI 0.37* MSCI 0.35* SML − 0.03 D-Ratio 0.001* T10Y 14.90 Credit Spread 1.63 Bd. Opt 0.07 FX Opt 0.16*** Com. Opt 0.05 Adj. R-squared 50.74% 50.93% Akaike info criterion (AIC) − 5.37 − 5.58 Schwarz info criterion (SIC) − 5.17 − 5.43 Estimation Technique OLS EGARCH Significance Levels: *** 10% and ** 5% and * 1%

Asia with Japan IndexDependent Variable: Asia with Japan Index return   Sample Period: January 2004 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept − 0.007** MSCI 0.26* MSCI 0.30* SML − 0.04 R-Ratio 2.81* T10Y 8.28 Kurtosis 0.006* Credit Spread 2.13 Bd. Opt − 0.05 FX Opt 0.11 Com. Opt 0.08** Adj. R-squared 34.64% 36.41% Akaike info criterion (AIC) − 5.20 − 5.54 Schwarz info criterion (SIC) − 5.00 − 5.36 Estimation Technique OLS EGARCH Significance Levels: *** 10% and ** 5% and * 1%

Japan IndexDependent Variable: Japan Index return   Sample Period: January 2004 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept 0.003** TOPIX 0.37* TOPIX 0.36* SML − 0.02 T10Y 26.87* Credit Spread 0.40 Bd. Opt 1.44 FX Opt 0.04 Com. Opt − 0.03 Adj. R-squared 64.12% 62.73% Akaike info criterion (AIC) − 5.56 − 5.57 Schwarz info criterion (SIC) − 5.38 − 5.53 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Equity Hedge StrategiesEmpirical Results

Equity HedgeDependent Variable: Equity Hedge return   Sample Period: January 1998 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.003 Intercept − 0.01* Intercept − 0.008* S&P500 0.30* S&P500 0.35* S&P500 0.32* SML 0.16* R-Ratio 3.71* R-Ratio 2.61* T10Y 15.48** Skew − 0.006* SML (Size Prem.) 0.16* Credit Spread − 0.16 L-Ratio − 0.28* Skew − 0.007* Bd. Opt 1.68** FX Opt 0.04 Com. Opt 0.10* Adj. R-squared 57.34% 55.24% 59.63% Akaike info criterion (AIC) − 5.46 − 5.62 − 5.70 Schwarz info criterion (SIC) − 5.31 − 5.49 − 5.57 Estimation Technique OLS GARCH GARCH Significance Levels: *** 10% and ** 5% and * 1%

EH: Energy/Basic MaterialsDependent Variable: EH: Energy/Basic Materials return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.001 Intercept 0.12* S&P500 0.21* S&P500 0.39* SML 0.09 Drawdown Velocity 0.001* T10Y 26.34 Run-up Velocity − 0.002* Credit Spread 2.98 Bd. Opt − 1.43 FX Opt − 0.03 Com. Opt 0.35* Adj. R-squared 60.38% 50.43% Akaike info criterion (AIC) − 4.65 − 4.59 Schwarz info criterion (SIC) − 4.46 − 4.45 Estimation Technique OLS ARCH Significance Levels: *** 10% and ** 5% and * 1%

EH: Equity Market NeutralDependent Variable: EH: Equity Market Neutral return   Sample Period: January 1998 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.003 Intercept 0.02* Intercept 0.0008 S&P500 − 0.02 Max Run-up% − 0.05* SML (Size Prem.) 0.06** SML 0.05** S 1 − 0.003* AR(1) 0.21* T10Y 1.96 L-Ratio 2.13* Credit Spread − 1.36 Bd. Opt − 0.13 FX Opt − 0.03 Com. Opt 0.01 Adj. R-squared 5.10% 7.27% 6.20% Akaike info criterion (AIC) − 6.16 − 6.27 − 6.20 Schwarz info criterion (SIC) − 6.01 − 6.13 − 6.15 Estimation Technique OLS EGARCH OLS Significance Levels: *** 10% and ** 5% and * 1%

EH: Fundamental GrowthDependent Variable: EH: Fundamental Growth return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.005 Intercept − 0.009* Intercept − 0.01* S&P500 0.26* S&P500 0.39* S&P500 0.40* SML − 0.01 D-Ratio 0.002* D-Ratio 0.001* T10Y 5.57 HML (Value Prem.) − 0.39* Credit Spread − 1.0 S 4 0.02* Bd. Opt − 1.21 FX Opt − 0.04 Com. Opt 0.16* Adj. R-squared 38.14% 43.53% 50.29% Akaike info criterion (AIC) − 4.61 − 4.79 − 4.85 Schwarz info criterion (SIC) − 4.42 − 4.67 − 4.73 Estimation Technique OLS ARCH OLS Significance Levels: *** 10% and ** 5% and * 1%

EH: Fundamental ValueDependent Variable: EH: Fundamental Value return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.0001 Intercept - 0.004*** Intercept − 0.004** S&P500 0.32* S&P500 0.38* S&P500 0.39* SML − 0.02 S 1 0.006* S 1 0.006* T10Y 27.44* AR(1) 0.23** HML (Value Prem.) − 0.17** Credit Spread − 0.04 AR(1) 0.23** Bd. Opt 0.22 FX Opt 0.07 Com. Opt 0.12* Adj. R-squared 62.95% 59.33% 60.62% Akaike info criterion (AIC) − 5.60 − 5.54 − 5.56 Schwarz info criterion (SIC) − 5.41 − 5.44 − 5.44 Estimation Technique OLS OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

EH: Multi-StrategyDependent Variable: EH: Multi-Strategy return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.004 Intercept − 0.002 Intercept − 0.001 S&P500 0.40* S&P500 0.55* S&P500 0.53 SML 0.23* S 1 0.005* S 1 0.005 T10Y 7.28 SML (Size Prem.) 0.24 Credit Spread 0.50 Bd. Opt − 3.39** FX Opt − 0.13 Com. Opt 0.12** Adj. R-squared 69.04% 61.77% 64.20% Akaike info criterion (AIC) − 5.04 − 5.01 − 5.07 Schwarz info criterion (SIC) − 4.85 − 4.87 − 4.90 Estimation Technique OLS GARCH GARCH Significance Levels: *** 10% and ** 5% and * 1%

EH: Quantitative DirectionalDependent Variable: EH: Quantitative Directional return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.008** Intercept 0.09* S&P500 0.14* S&P500 0.20* SML 0.15** Max Run-up% − 0.09* T10Y 6.53 Drawdown Velocity − 0.001* Credit Spread − 2.65*** S 3 5.66* Bd. Opt 0.91 Std. Dev. − 2.98* FX Opt 0.07 Com. Opt 0.07 Adj. R-squared 24.23% 39.00% Akaike info criterion (AIC) − 5.31 − 5.54 Schwarz info criterion (SIC) − 5.12 − 5.40 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

EH: Short BiasDependent Variable: EH: Short Bias return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept 0.02* Intercept 0.005** S&P500 − 0.56* S&P500 − 0.62* S&P500 − 0.51* SML − 0.46* Run-up Duration − 0.002* SML − 0.48* T10Y − 4.51 Max Drawdown% − 0.08* Drawdown Velocity 0.0003* Credit Spread − 2.12 Sharpe Ratio 0.02* Bd. Opt 0.16 FX Opt − 0.06 Com. Opt 0.01 Adj. R-squared 77.06% 69.41% 78.83% Akaike info criterion (AIC) − 5.38 − 5.29 − 5.49 Schwarz info criterion (SIC) − 5.19 − 5.13 − 5.40 Estimation Technique OLS ARCH OLS Significance Levels: *** 10% and ** 5% and * 1%

EH: Technology/HealthcareDependent Variable: EH: Technology/Healthcare return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.007** Intercept − 0.07* Intercept 0.02*** S&P500 0.26* S&P500 0.33* S&P500 0.29* SML 0.11 Sortino Ratio 0.03* HML (Value Prem.) − 0.29* T10Y 15.18 S 4 − 0.03* Sortino Ratio 0.01* Credit Spread − 0.66 Kurtosis − 0.008* Run-up Velocity − 0.001** Bd. Opt 0.03 Std. Dev. 3.23* SML (Size Prem.) 0.20* FX Opt 0.05 Com. Opt 0.02 Adj. R-squared 39.26% 45.46% 52.39% Akaike info criterion (AIC) − 5.37 − 5.66 − 5.69 Schwarz info criterion (SIC) − 5.18 − 5.45 − 5.48 Estimation Technique OLS GARCH GARCH Significance Levels: *** 10% and ** 5% and * 1%

Relative Value StrategiesEmpirical Results

Relative Value ArbitrageDependent Variable: Relative Value Arbitrage return   Sample Period: January 1998 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.007 Intercept − 0.003 S&P500 0.09* S&P500 0.10* SML 0.02 R-Ratio 1.79* T10Y 11.45*** AR(1) 0.46* Credit Spread − 2.74 Bd. Opt − 0.48 FX Opt 0.03 Com. Opt 0.08* Adj. R-squared 47.55% 42.86% Akaike info criterion (AIC) − 5.59 − 6.08 Schwarz info criterion (SIC) − 5.45 − 5.97 Estimation Technique OLS GARCH Significance Levels: *** 10% and ** 5% and * 1%

RV: Energy InfrastructureDependent Variable: RV: Energy Infrastructure return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.009** Intercept − 0.10* Intercept − 0.01 S&P500 0.30* S&P500 0.38* S&P500 0.39* SML − 0.12 R-Ratio 6.73* HML (Value Prem.) − 0.47* T10Y 7.66 Std. Dev. 2.56* R-Ratio 5.14* Credit Spread 0.03 Drawdown Velocity 0.001* Bd. Opt − 2.54 Max Drawdown% − 0.10* FX Opt 0.16 Com. Opt 0.24* Adj. R-squared 49.15% 41.90% 48.45% Akaike info criterion (AIC) − 4.76 − 4.66 − 4.76 Schwarz info criterion (SIC) − 4.57 − 4.57 − 4.62 Estimation Technique OLS OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

RV: Fixed Income–Asset BackedDependent Variable: RV: Fixed Income–Asset Backed return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.009* Intercept − 0.02* S&P500 0.08* Drawdown Duration 0.004* SML − 0.03 S&P500 0.05* T10Y 0.21 S 4 0.15* Credit Spread 0.97 AR(1) 0.37* Bd. Opt 0.13 FX Opt 0.11** Com. Opt 0.06** Adj. R-squared 30.48% 41.46% Akaike info criterion (AIC) − 6.49 − 6.70 Schwarz info criterion (SIC) − 6.25 − 6.56 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

RV: Fixed Income–Convertible ArbitrageDependent Variable: RV: Fixed Income–Convertible Arbitrage return   Sample Period: January 1998 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.01 Intercept − 0.001 S&P500 0.12* S&P500 0.06* SML 0.01 R-Ratio 3.93* T10Y − 13.58*** AR(1) 0.77* Credit Spread − 7.68 Bd. Opt − 0.50 FX Opt 0.09 Com. Opt 0.14* Adj. R-squared 43.37% 37.50% Akaike info criterion (AIC) − 4.53 − 5.65 Schwarz info criterion (SIC) − 4.38 − 5.54 Estimation Technique OLS GARCH Significance Levels: *** 10% and ** 5% and * 1%

RV: Fixed Income–CorporateDependent Variable: RV: Fixed Income–Corporate return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.008** Intercept 0.004* S&P500 0.17* S&P500 0.19* SML − 0.002 T10Y 9.25 Credit Spread − 3.11*** Bd. Opt 0.42 FX Opt 0.11*** Com. Opt 0.12* Adj. R-squared 54.45% 45.25% Akaike info criterion (AIC) − 5.84 − 6.12 Schwarz info criterion (SIC) − 5.62 − 5.98 Estimation Technique OLS GARCH Significance Levels: *** 10% and ** 5% and * 1%

RV: Fixed Income–SovereignDependent Variable: RV: Fixed Income–Sovereign return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.002 Intercept − 0.003 S&P500 0.22* S&P500 0.23* SML − 0.04 R-Ratio 0.91** T10Y − 1.73 Credit Spread 1.87 Bd. Opt 1.93 FX Opt 0.19*** Com. Opt 0.22* Adj. R-squared 38.07% 27.32% Akaike info criterion (AIC) − 5.15 − 5.66 Schwarz info criterion (SIC) − 4.96 − 5.52 Estimation Technique OLS GARCH Significance Levels: *** 10% and ** 5% and * 1%

RV: Multi-StrategyDependent Variable: RV: Multi-Strategy return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept 0.02* S&P500 0.13* S&P500 0.18* SML 0.001 Max Run-up% − 0.06* T10Y 4.62 Drawdown Duration 0.001* Credit Spread 1.50 Bd. Opt − 0.80 FX Opt − 0.004 Com. Opt 0.10* Adj. R-squared 48.15% 46.39% Akaike info criterion (AIC) − 5.82 − 6.33 Schwarz info criterion (SIC) − 5.61 − 6.12 Estimation Technique OLS GARCH Significance Levels: *** 10% and ** 5% and * 1%

RV: Real EstateDependent Variable: RV: Real Estate return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.004 Intercept 0.02** Intercept 0.002 S&P500 0.29* S&P500 0.31* S&P500 0.30* SML 0.18* Max Run-up% − 0.11* SML (Size Prem.) 0.12** T10Y 7.90 L-Ratio 0.97** Credit Spread 3.07 Max Drawdown% − 0.07* Bd. Opt 0.36 FX Opt 0.09 Com. Opt 0.09* Adj. R-squared 56.83% 54.98% 54.00% Akaike info criterion (AIC) − 5.61 − 5.74 − 5.68 Schwarz info criterion (SIC) − 5.39 − 5.55 − 5.52 Estimation Technique OLS GARCH GARCH Significance Levels: *** 10% and ** 5% and * 1%

RV: VolatilityDependent Variable: RV: Volatility return   Sample Period: January 2004 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.004 Intercept − 0.002 S&P500 0.07 D-Ratio 0.001* SML 0.01 L-Ratio − 0.28** T10Y 24.73* Credit Spread − 0.98 Bd. Opt 0.72 FX Opt 0.07 Com. Opt − 0.04 Adj. R-squared 8.05% 10.30% Akaike info criterion (AIC) − 5.62 − 5.70 Schwarz info criterion (SIC) − 5.42 − 5.63 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

RV: Yield AlternativeDependent Variable: RV: Yield Alternative return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.005 Intercept 0.003** S&P500 0.25* S&P500 0.34* SML − 0.006 S 1 0.01* T10Y 2.40 L-Ratio − 0.60* Credit Spread 1.07 Bd. Opt − 2.85** FX Opt 0.21*** Com. Opt 0.21* Adj. R-squared 44.62% 42.14% Akaike info criterion (AIC) − 4.84 − 5.23 Schwarz info criterion (SIC) − 4.63 − 5.07 Estimation Technique OLS GARCH Significance Levels: *** 10% and ** 5% and * 1%

Event Driven StrategiesEmpirical Results

Event DrivenDependent Variable: Event Driven return   Sample Period: January 1998 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.003 Intercept 0.003** Intercept − 0.02 S&P500 0.22* S&P500 0.25* S&P500 0.24 SML 0.14* S 5 0.61* SML (Size Prem.) 0.17 T10Y 18.56* AR(1) 0.33* Max Run-up% − 0.01 Credit Spread 0.14 S 5 0.71 Bd. Opt 0.43 Drawdown Velocity − 0.001 FX Opt 0.06 Com. Opt 0.05** Adj. R-squared 55.92% 49.52% 53.84% Akaike info criterion (AIC) − 5.84 − 5.80 − 5.93 Schwarz info criterion (SIC) − 5.69 − 5.68 − 5.80 Estimation Technique OLS GARCH ARCH Significance Levels: *** 10% and ** 5% and * 1%

ED: ActivistDependent Variable: ED: Activist return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.006 Intercept − 0.003 S&P500 0.60* S&P500 0.74* SML 0.22** S 4 0.02* T10Y 19.95 Credit Spread 5.96* Bd. Opt 0.84 FX Opt − 0.05 Com. Opt 0.18* Adj. R-squared 71.50% 66.37% Akaike info criterion (AIC) − 4.74 − 4.65 Schwarz info criterion (SIC) − 4.55 − 4.50 Estimation Technique OLS GARCH Significance Levels: *** 10% and ** 5% and * 1%

ED: Credit ArbitrageDependent Variable: ED: Credit Arbitrage return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.004* Intercept 0.006* S&P500 0.10* S&P500 0.11* SML 0.01 Skew 0.01* T10Y 24.10* S 5 0.45** Credit Spread 1.27*** AR(1) 0.31* Bd. Opt − 0.16 FX Opt − 0.07 Com. Opt 0.03 Adj. R-squared 55.41% 47.10% Akaike info criterion (AIC) − 6.84 − 6.70 Schwarz info criterion (SIC) − 6.65 − 6.58 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

ED: Distressed/RestructuringDependent Variable: ED: Distressed/Restructuring return   Sample Period: January 1998 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.02* Intercept − 0.003 Intercept − 0.005*** S&P500 0.14* S&P500 0.16* S&P500 0.17* SML 0.10* R-Ratio 2.06* R-Ratio 1.95* T10Y 8.04 AR(1) 0.34* SML (Size Prem.) 0.09* Credit Spread − 8.70* AR(1) 0.39* Bd. Opt − 0.57 FX Opt 0.04 Com. Opt 0.06** Adj. R-squared 44.77% 36.69% 38.75% Akaike info criterion (AIC) − 5.48 − 5.52 − 5.59 Schwarz info criterion (SIC) − 5.33 − 5.40 − 5.44 Estimation Technique OLS GARCH EGARCH Significance Levels: *** 10% and ** 5% and * 1%

ED: Merger ArbitrageDependent Variable: ED: Merger Arbitrage return   Sample Period: January 1998 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.001 Intercept 0.006* Intercept 0.006* S&P500 0.09* S&P500 0.09* S&P500 0.09* SML 0.05* S 5 4.93* S 5 4.85* T10Y 7.37 S 3 − 5.91** S 3 − 6.23* Credit Spread 1.88** SML (Size Prem.) 0.04** Bd. Opt 0.85*** FX Opt 0.008 Com. Opt 0.03** Adj. R-squared 23.27% 30.65% 31.98% Akaike info criterion (AIC) − 6.57 − 6.69 − 6.70 Schwarz info criterion (SIC) − 6.44 − 6.62 − 6.62 Estimation Technique OLS OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

ED: Multi-StrategyDependent Variable: ED: Multi-Strategy return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.002 Intercept − 0.004 Intercept − 0.005** S&P500 0.33* S&P500 0.37* S&P500 0.39* SML − 0.14 S 5 2.01* S 5 2.18* T10Y 30.70** HML (Value Prem.) − 0.26** Credit Spread 0.80 Bd. Opt 1.59 FX Opt 0.14 Com. Opt 0.18* Adj. R-squared 46.48% 43.48% 45.92% Akaike info criterion (AIC) − 4.96 − 4.94 − 4.98 Schwarz info criterion (SIC) − 4.77 − 4.87 − 4.88 Estimation Technique OLS OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

ED: Special SituationsDependent Variable: ED: Special Situations return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept − 0.0004 Intercept − 0.002 Intercept − 0.001 S&P500 0.29* S&P500 0.32* S&P500 0.32* SML 0.02 S 1 0.007* HML (Value Prem.) − 0.31* T10Y 21.85** AR(1) 0.32* S 1 0.007* Credit Spread 0.58 AR(1) 0.27** Bd. Opt 1.06 FX Opt 0.15** Com. Opt 0.12* Adj. R-squared 57.64% 53.53% 59.09% Akaike info criterion (AIC) − 5.54 − 5.49 − 5.68 Schwarz info criterion (SIC) − 5.33 − 5.39 − 5.51 Estimation Technique OLS OLS ARCH Significance Levels: *** 10% and ** 5% and * 1%

Europe Regional StrategiesEmpirical Results

Western/Pan Europe IndexDependent Variable: Western /Pan Europe Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.006 Intercept − 0.03* MSCI 0.32* MSCI 0.20* SML − 0.13*** R-Ratio 8.73* T10Y 8.85 Max Drawdown% − 0.26* Credit Spread − 1.21 L-Ratio − 0.42* Bd. Opt − 0.27 Max Run-up% − 0.06* FX Opt 0.31* Com. Opt 0.06 Adj. R-squared 39.66% 57.24% Akaike info criterion (AIC) − 5.25 − 5.62 Schwarz info criterion (SIC) − 5.04 − 5.48 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Northern Europe IndexDependent Variable: Northern Europe Index return   Sample Period: January 2005 – December 2014     FH 7-Factor Model CAPM Model Fama -French Model Variable Coefficient Variable Coefficient Variable Coefficient Intercept 0.0003 Intercept 0.003* MSCI 0.18* MSCI 0.12* SML 0.003 T10Y 7.68 Credit Spread 1.62 Bd. Opt 0.79 FX Opt 0.30* Com. Opt 0.06*** Adj. R-squared 28.46% 17.73% Akaike info criterion (AIC) − 6.03 − 5.94 Schwarz info criterion (SIC) − 5.84 − 5.89 Estimation Technique OLS OLS Significance Levels: *** 10% and ** 5% and * 1%

Final Remarks55 HFRX categories of hedge fund returns are analyzed. Three new factors (D-Ratio, L-Ratio, and R-Ratio) are introduced as better risk measures for hedge fund categories than traditional factors. The three new factors are dominant in the Macro category, the Relative Value category, the Emerging Market category, and the Asia regional categories (at least one of these variables is significant in 74% of these models). These three variables are not as dominant in the Equity Hedge category (33%) or the Event Driven category (14%).L-Ratio is significant in 12 of 55 categories (21.8%).R-Ratio is significant in 17 of 55 categories (30.9%).D-Ratio is significant in 13 of 55 categories (23.6%).The Sharpe and Sortino ratios are only significant in 4-7 categories of the 55 hedge fund groups.