Three High-Dimensional Cross-Sectional Tests for Fund Performance Evaluation

48 Pages Posted: 8 Jul 2019

See all articles by Tingting Cheng

Tingting Cheng

Monash University

Cheng Yan

University of Essex - Essex Business School

Yayi Yan

Department of Econometrics and Business Statistics

Date Written: May 7, 2019

Abstract

The traditional fund-by-fund performance evaluation method suffers from various econometric problems such as multiple hypothesis testing, time-varying coefficients, cross-sectional dependence, etc. To overcome these problems, we tailor three high-dimensional cross-sectional tests to empirically evaluate the performance of both U.S. mutual funds and hedge funds, which yield results fundamentally different from the extant literature. For mutual (hedge) funds, we find about 5% (10%) on average with significant alphas, and more negative (positive) alphas than the positive (negative) alphas. The statistical significance of these fund alphas is too high to be explained by sheer luck. Interestingly, fund performance diverges during the late 2000s Global Financial Crisis. In addition, we find some (little) arbitrage opportunities among mutual (hedge) funds.

Keywords: Performance Evaluation, Mutual Fund, Hedge Fund, Crisis, Arbitrage

JEL Classification: C15, G11, G12, G23

Suggested Citation

Cheng, Tingting and Yan, Cheng and Yan, Yayi, Three High-Dimensional Cross-Sectional Tests for Fund Performance Evaluation (May 7, 2019). Available at SSRN: https://ssrn.com/abstract=3383867 or http://dx.doi.org/10.2139/ssrn.3383867

Tingting Cheng

Monash University ( email )

23 Innovation Walk
Wellington Road
Clayton, Victoria 3800
Australia

Cheng Yan (Contact Author)

University of Essex - Essex Business School ( email )

Wivenhoe Park
Colchester, CO4 3SQ
United Kingdom

Yayi Yan

Department of Econometrics and Business Statistics ( email )

900 Dandenong Road
Caulfield, Victoria 3145
Australia
0498101110 (Phone)

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
52
Abstract Views
311
rank
400,341
PlumX Metrics