Detecting Performance Persistence in Fund Managers: Book Benchmark Alpha Analysis

Preprint
Publisher Website
Google Scholar
Abstract
We develop a new approach for relative evaluation of fund managers within a portfolio (Book). Based on the explicit positions of the funds and the positions of the overall portfolio, we decompose each fund's return into a beta and alpha component relative to the overall Book. We test this Book Benchmark analysis on a portfolio of equity-based hedge funds during a 31 month period. Our results indicate that Book Benchmark alphas are significantly more predictive than returns when the in-sample periods are short (six to nine months). This suggests that Book Benchmark alphas are a valuable quantitative tool for managing a portfolio of hedge funds with position level transparency. While our analysis is developed in the context of a fund of hedge funds because of data considerations, the Book Benchmark analysis is more general. It can be used in any situation involving manager selection as long as there is position level transparency.

This publication has 11 references indexed in Scilit: