Data Sourcing, Statistical Processing and Time Series Analysis Presented at EDAMBA summer school, Soréze (France) 23 July – 27 July 2009 An Example from Research into Hedge Fund Investments
Presenter:Florian Boehlandt University:University of Stellenbosch – Business School Supervisor:Prof Eon Smit Prof Niel Krige Research Title:Pricing hedge funds a.k.a. The sustainability of parametric and semi- parametric pricing models as estimators of hedge fund performance
‘In the business world, the rearview mirror is always clearer than the windshield’ - Warren Buffett -
Research Purpose 1.Developing accurate parametric pricing models for hedge funds and fund of hedge funds 2.Accounting for the special statistical properties of alternative investment funds 3.Providing practitioners and statisticians with a framework to assess, categorize and predict hedge fund investments
Research Approach Logical-positivistic, deductive research: Postulation of hypotheses that are tested via standard statistical procedures Research Philosophy Empirical analysis: Interpreting the quality of pricing models on the basis of historical data Research Approach External secondary data: Historic time series adjusted for data-bias effects Data Sourcing
Data Sources Hedge Fund Databases CISDM/MAR Financial Databases Risk Simulation Monte Carlo (Solver) Confidence (RiskSim) Data Sourcing DATA POOL
FACTOR ANALYSIS Data Treatment Risk Simulation Statistical Processing Excel / VBA Statistica EViews Data Treatment DATA POOL MODEL BUILDING STATISTICAL CLUSTERING
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
Database Management Avoiding duplicate entries Cross-referencing data from various sources Combining and aggregating different databases Efficient storage due to relational data management Queries allow for retrieval/display of specific data Linked-in with Microsoft VBA and Excel (data displayable as Pivot table reports) Searching for specific entries via SQL
Data Bias Survivorship Self- Selection Database Instant History Look-ahead Inclusion of graveyard funds Multiple databases Rolling-window observation / Incubation period
Statistical tests for TSA Regression Statistics (Alpha, Average Error term, Information Ratio) Normality (Chi-squared, Jarque Bera) Goodness of fit, phase-locking and collinearity (Akaike Information Criterion, Hannan-Schwartz) Serial Correlation (Durbin-Watson, Portmanteau) Non-stationarity (unit root)
Prediction Models AR ARMA ARIMA GLS Univariate Multivariate Conditional PCA Polynomial Fitting Taylor Series Higher Co- Moments Constrained Lagrange KKT Simulation
Literature Review 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)
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: