Empirical Analysis of Fund of Hedge Funds (Tass database) Presented to: Research Project and Working Paper
Presenter:Florian Boehlandt University:University of Stellenbosch – Business School Supervisor:Prof Eon Smit Prof Niel Krige Research Title:A Risk-Return Assessment of Fund of Hedge Funds in Comparison to Single Hedge Funds – An Empirical Analysis
‘In the business world, the rearview mirror is always clearer than the windshield’ - Warren Buffett -
Research Purpose 1.Comparative time series analysis of Fund of Hedge Funds vs. Single Manager Funds 2.Estimating the impact of leverage on downside volatility and risk 3.Constructing style indices from risk parameters and AUM weightings 4.Automating data import and data analysis for future quantitative analysis (‘dashboard’)
Code Execution (1/2) Data Import Extract relevant data from Access (SQL) Import data as Pivot table report Data Treatment Test for serial correlation / normality Calculate adjusted excess returns Data Analysis Select funds with consistent data series Determine fund specific risk parameters
Code Execution (2/2) Weighting Estimate weighted average parameters Construct style indices Comparative Analysis Calculate within-group variation Calculate between-group variation Data Output Tabular display of aggregate results Construction of line - bar charts
Hedge Fund Categories (TASS) Categories Directional Dedicated Short Bias Global Macro Emerging Markets Global Macro Long / Short Equity Managed Futures Fund of Hedge Funds Market Neutral Equity Market Neutral Event Driven Convertible Arbitrage Fixed Income Arbitrage
Data Import Code Fund (Name) Main Strategy Information MM_DD_YYYY (Date) Yield Ptype (ROI or AUM) Performance Leverage (Yes/No) System Information Access DatabaseExcel Pivot table report
Risk-Return Parameters (1/2) Return on Investment Downside Risk – Standard Deviation – Downside Deviation – Value at Risk – Modified Value at Risk Maximum Continuous Drawdown
Risk-Return Parameters (2/2) 3-Factor Regression – Regression Alpha – Average Error term – Information Ratio Adaptation Current Research
t – test (between strategies) Unbalanced ANOVA (within and between treatments) t – test (leverage vs. no leverage) t – test for equal means t – test for equal means t – test for equal means Statistical Tests Strategy 1 Leverage Strategy 1 No Leverage t – test for equal means Strategy 2 Leverage Strategy 2 No Leverage
Step 1: Copy folder to desktop or hard drive User Guide (1/4)
Step 2: Manual amendments to source code User Guide (2/4) '*********************************************** '-->RAWDATA() 'Rotate through PivotItems Strategy '1 = Convertible Arbitrage '2 = Dedicated Short Bias '3 = Emerging Markets '4 = Equity Market Neutral '5 = Event Driven '6 = Fixed Income Arbitrage '7 = Fund of Funds '8 = Global Macro '9 = Long/Short Equity '10 = Managed Futures '*********************************************** 'set index and endloopi to stategy in focus (e.g. 7 for Fund of Funds) 'set endloopj and startloopj to strategies compared 'e.g. comparing Fund of Hedge Funds to Fixed Income Arbitrage : index = 7 endloopi = 7 endloopj = 7 startloopj = 6
Step 3: Open spreadsheet shell and start execution User Guide (3/4)
Step 4: Fill in Userform User Guide (4/4) Select hard drive Select file path Select parameter
Example Output (1/2)
Example Output (2/2) Joint starting point Significance
Empirical Findings (1/2) Measures of volatility and downside risk were significantly improved for FoHFs, compared to their single-strategy peers No evidence was found that FoHF strategies overcharge for risk diversification benefits With reference to continuous drawdown, attrition rates and VaR, FoHFs are a valuable supplement to the institutional portfolio
Empirical Findings (2/2) It could not be established whether gearing affected hedge fund performance – either favourably or adversely Some statistical evidence could be found of a higher exposure of leveraged funds to the recent subprime crisis
Extended Research Hedge Fund Linear Pricing Models – Sharpe Factor Model (Sharpe, 1992) – Constrained Regression (Otten, 2000) – Fama-French Factor Model (Fama, 1992) Factor Component Analysis (Fung, 1997) Simulation of Trading component (lookback straddle)
Prediction Models AR ARMA ARIMA GLS Univariate Multivariate Conditional PCA Polynomial Fitting Taylor Series Higher Co- Moments Constrained Lagrange KKT Simulation
Sources Fama, E.F. & French, K.R The Cross-Section of Expected Stock Returns. Journal of Finance, 47(2), June, [Online] Available: % %2947%3A2%3C427%3ATCOESR%3E2.0.CO%3B2-N % %2947%3A2%3C427%3ATCOESR%3E2.0.CO%3B2-N Fung, W. & Hsieh, D.A Empirical characteristics of dynamic trading strategies: the case of hedge funds. Review of Financial Studies, 10(2), Summer, [Online] Available: Otten, R. & Bams, D Statistical Tests for Return-Based Style Analysis. Paper delivered at EFMA 2001 Lugano Meetings, July. [Online] Available: Sharpe, W.F Asset allocation: management style and performance measurement. Journal of Portfolio Management, Winter, [Online] Available: