Time-Varying Style Analysis and the Leverage of Funds of Hedge Funds

Posted: 3 Oct 2012 Last revised: 18 Mar 2013

See all articles by Benoit Dewaele

Benoit Dewaele

Université Libre de Bruxelles (ULB)

Date Written: October 2, 2012

Abstract

20 years ago, Sharpe (1992) developed the Style Analysis for mutual funds; in this analysis, the weights mutual funds allocate to major asset classes are constrained to sum up to 1. In this paper we develop a Time-Varying Style Analysis (TVSA) in which the weights must sum up to 1 but are allowed to vary with time, and show its interest in the world of Funds of Hedge Funds (FoHFs). We show that this approach sheds light on the leverage utilized by funds of hedge funds and subsequently use this estimate to determine the impact of leverage on future performance. Using predictive multivariate analysis, we show that the impact of leverage is negative, effect that, to the best of our knowledge, has never been reported. We also show that TVSA leads to lower median tracking error than the classical style analysis and allow us to determine an estimate of the aggregate leverage of the FoHF industry. Moreover, applying Darolles and Vaissié (2011) decomposition to our results confirms that funds of hedge fund picking skills are not enough to compensate for the fees they charge. Our results prove to be robust to adjustments for backfilling, survivorship and autocorrelation bias and to various robustness checks.

Keywords: hedge funds, fund of hedge funds, time-varying coefficients models, leverage

JEL Classification: C22, G11, G23

Suggested Citation

Dewaele, Benoit, Time-Varying Style Analysis and the Leverage of Funds of Hedge Funds (October 2, 2012). Available at SSRN: https://ssrn.com/abstract=2155618 or http://dx.doi.org/10.2139/ssrn.2155618

Benoit Dewaele (Contact Author)

Université Libre de Bruxelles (ULB) ( email )

CP 132 Av FD Roosevelt 50
Brussels, Brussels 1050
Belgium

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views
473
PlumX Metrics