Improved Inference in the Evaluation of Mutual Fund Performance Using Panel Bootstrap Methods

Abstract

This is the accepted version of the paper.This version of the publication may differ from the final published version. Permanent repository link AbstractTwo new methodologies are introduced to improve inference in the evaluation of mutual fund performance against benchmarks. First, the benchmark models are estimated using panel methods with both fund and time effects. Second, the non-normality of individual mutual fund returns is accounted for by using panel bootstrap methods. We also augment the standard benchmark factors with fund-specific characteristics, such as fund size. Using a dataset of UK equity mutual fund returns, we find that fund size has a negative effect on the average fund manager's benchmark-adjusted performance. Further, when we allow for time effects and the non-normality of fund returns, we find that there is no evidence that even the best performing fund managers can significantly out-perform the augmented benchmarks after fund management charges are taken into account.

A Time-Series Bootstrapping Simulation Method to Distinguish Sell-Side Analysts’ Skill from Luck

“…Specifically, one simulation method recognizes the panel nature of the dataset and the existence of both fund and time effects, while the other one takes into account the non-normal distribution of individual mutual fund returns. Blake et al (2014) find little evidence that UK fund managers can generate superior abnormal returns when the non-normality of fund returns using a series of both parametric and non-parametric bootstrapping simulations is allowed. Barras, Scaillet, and Wermers (2010) propose an alternative model to distinguish skilled (unskilled) funds from lucky (unlucky) funds.…”

“…They find that the top hedge fund managers possess certain asset selection skill and can take advantage of the simple trading strategies, the results of which challenge the classical point of view that the top hedge funds are just lucky and their performance could not be persistent. Blake, Caulfield, Ioannidis, and Tonks (2014) argue that the non-parametric bootstrapping simulation methods of Kosowski et al (2006) and Fama and French (2010) are flawed as the returns are drawn from a uniform distribution. 1 To alleviate the limitations of the non-parametric bootstrapping simulations, Blake et al, (2014) suggest the application of the parametric bootstrapping simulations, in which the time-series returns for each fund are resampled from a stable distribution that shows the distributional properties of the realized returns over the whole sample period.…”

“…Blake, Caulfield, Ioannidis, and Tonks (2014) argue that the non-parametric bootstrapping simulation methods of Kosowski et al (2006) and Fama and French (2010) are flawed as the returns are drawn from a uniform distribution. 1 To alleviate the limitations of the non-parametric bootstrapping simulations, Blake et al, (2014) suggest the application of the parametric bootstrapping simulations, in which the time-series returns for each fund are resampled from a stable distribution that shows the distributional properties of the realized returns over the whole sample period. For example, to identify the underlying distribution of alphas, Meyer, Schmoltzi, Stammschulte, Kaesler, Loos, and Hackethal (2012) draw a new alpha that is added to the return series for each portfolio and for each simulation run.…”

“…An improper application of the block bootstrapping simulation method could underestimate the risk of a portfolio with the independent time-varying returns and overestimate the risk of a portfolio with the mean-reverting returns. Blake et al (2014) introduce two new simulation methods allowing for appropriate inference in the presence of the non-normal asset returns when evaluating the performance of mutual funds against the benchmark model. Specifically, one simulation method recognizes the panel nature of the dataset and the existence of both fund and time effects, while the other one takes into account the non-normal distribution of individual mutual fund returns.…”