Tactical Asset Allocation on Market Neutral Hedge Fund

Abstract:     
The objective of the thesis is to show through an empirical work how alpha drivers can be used tactically with beta drivers to provide solid out-performance compared to a chosen benchmark. Given the fact that financial theory and empirical research cast doubt on the alpha generating process based on stock-picking abilities by Fund Manager, I substitute that methodology with a quantitative approach. Using a robust econometric process based on a non-linear multi-factor thick and recursive modeling approach that takes into account structural breaks in the data generating process, I found statistically and economically significant evidence of returns predictability for the DJ Euro Stoxx 50 excess returns. My modeling approach is capable of accounting for model specification uncertainty, possible shifts in the forecasting model and low probability events. Based on this predictability poIr, I back test the implementation of Asset Allocation Strategies simulating three Market Neutral Hedge Funds on different asset classes and degree of risk aversion. I back test these portfolios trading Index futures to generate alpha drivers, and an optimized option strategy to generate beta drivers. Transaction, administration and management fees are included in order to be very close to market's real conditions. As a conclusion, out-performance on risk-adjusted returns is obtained through the implementation of a dynamic, robust and complete econometric process, during periods of downward and upward markets.

Number of Pages in PDF File: 51

Keywords: Hedge Fund, Alpha, Beta, Multi factor thick model, econometrics

JEL Classification: C50, G10

working papers series