A Stepwise SPA Test for Data Snooping and its Application on Fund Performance Evaluation

31 Pages Posted: 28 Feb 2006  

Po-Hsuan Hsu

University of Hong Kong

Yu-Chin Hsu

University of Missouri at Columbia - Department of Economics

Date Written: March 31, 2006

Abstract

In this study, we propose a Stepwise SPA Test which is powerful in searching for predictive models or profitable investment targets with appropriate family-wise error control. Our testing method, built on White's Reality Check (2000), Hansen's SPA test (2005), and Romano and Wolf's stepwise procedure (2005), aims to conduct large-scale joint hypothesis testing in a given data set. Based on the Monte Carlo simulations, we show that the proposed test is more powerful than the other three tests in limited samples. We then apply the test to examine the performance of mutual funds and hedge funds, and obtain some interesting results. We find that only eight mutual funds are found to beat the S&P 500 index, and only few hedge funds outperform the risk-free rate. With these two empirical cases, we substantiate the empirical value of our test in fund performance evaluation.

Keywords: Data Snooping, Reality Check, Stepwise Test, SPA test

JEL Classification: C12, C32

Suggested Citation

Hsu, Po-Hsuan and Hsu, Yu-Chin, A Stepwise SPA Test for Data Snooping and its Application on Fund Performance Evaluation (March 31, 2006). Available at SSRN: https://ssrn.com/abstract=885364 or http://dx.doi.org/10.2139/ssrn.885364

Po-Hsuan Hsu

University of Hong Kong ( email )

Pokfulam Road
Hong Kong
China

Yu-Chin Hsu (Contact Author)

University of Missouri at Columbia - Department of Economics ( email )

118 Professional Building
Columbia, MO 65211
United States
(573) 882-6474 (Phone)
(573) 882-2697 (Fax)

HOME PAGE: http://yuchinhsu.yolasite.com/

Paper statistics

Downloads
425
Rank
54,104
Abstract Views
3,061