Data Mining Corrections Testing in Chinese Stocks

Abstract

In this analysis of the risk and return of stocks in global and Chinese markets, we build a reasonably large number of models for stock selection and create optimized portfolios to outperform a global benchmark. We apply robust regression techniques in producing stock-selection models and Markowitz-based optimization techniques in portfolio construction in a global stock universe and two Chinese stock universes. We report the results of applying a data mining corrections test to the global and Chinese stock universes. We find that (1) robust regression applications are appropriate for modeling stock returns in global and Chinese stock markets; (2) mean-variance techniques continue to produce portfolios capable of generating returns that exceed transactions costs; and (3) our global portfolio selection models pass data mining tests, such that the models produce statistically significant asset selection for global and MSCI-China universes, but not for China A-shares.