Evaluating hedge fund managers: A Bayesian investigation of skill and persistence

Vrontos Ioannis (), Vrontos Spyridon and Daniel Giamouridis
Additional contact information
Vrontos Ioannis: Athens University of Economics and Business
Vrontos Spyridon: University of Aegean

No 487, Computing in Economics and Finance 2006 from Society for Computational Economics

Abstract: The last couple of decades has witnessed a growing interest in hedge funds. Academics and practitioners are intrigued by the distinct characteristics of these investment vehicles: hedge funds are flexible with respect to the types of securities they hold and the type of positions they take; they are not subject to public disclosure of their activities; and they are not evaluated against a passive benchmark. Operating in this environment encourages hedge fund managers to construct highly dynamic trading strategies and as a result expose their portfolios to a plethora of economic risk factors. From the investor point of view, very little is known about the investment process, the economic risks associated with a particular hedge fund or a hedge fund strategy, and the skills of the hedge fund manager. While the details of the investment process could be of little practical importance to the investor, the latter – risks and managerial skill - are not. The objective of separating ‘alpha’ and ‘beta’ - the non-systematic and systematic components of returns respectively – is thus one of the main themes of current interest on hedge fund investing. This work introduces a Bayesian approach to estimate the ‘alpha’ and ‘beta’ of hedge fund investments. The Bayesian analysis consists of parameter estimation and model selection, and is implemented via Markov Chain Monte Carlo methodologies. The proposed stochastic search algorithm is appealing in identifying the relevant systematic risk factors in situations where the set of possible factors is very large, i.e. in hedge funds (‘black box’ investments). The entire analysis is carried out in a dynamic setup where hedge fund return volatilities change over time. We address two empirical questions. First we examine if the proposed methodology has an impact on the relative ranking of hedge fund managers compared with current approaches. Second, we study the nature of persistence in hedge fund return performance, i.e. short- or long-term. These questions are extremely interesting for the practice of hedge fund investing. A reliable ‘alpha’ estimation leads to more efficient manager selection schemes, i.e. fund of hedge funds or multi-strategy funds. In addition identifying the nature of persistence in hedge fund investment performance helps in determining successful portfolio rebalancing strategies (timing).

Keywords: Bayesian methods; Hedge Funds; Markov Chain Monte Carlo; Manager skill (search for similar items in EconPapers)
JEL-codes: C11 G12 (search for similar items in EconPapers)
Date: 2006-07-04
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sce:scecfa:487

Access Statistics for this paper

More papers in Computing in Economics and Finance 2006 from Society for Computational Economics Contact information at EDIRC.
Bibliographic data for series maintained by Christopher F. Baum ().