Testing and Filtering Quantitative Factors in China’s A-Share Market

Summary

This report evaluates 56 equity factors built from Chinese A-share trading and financial statement data. It sorts stocks into ten groups by each factor, measures the next month’s value-weighted portfolio returns, and tests the return spread between the extreme groups. It covers the full sample from 1997 to 2017 and a post-share-reform sample from 2007 to 2017, reporting return, volatility, Sharpe ratio, drawdown, and statistical tests. The authors find 13 effective factors in the full sample and 10 in the later period, with trading-friction and liquidity-related measures prominent in both.

The report also examines correlations among factors and discusses removing redundant measures within factor categories. It warns that orthogonalization can depend heavily on the order in which factors are introduced. The findings are specific to the study’s sample, data sources, portfolio construction, and testing choices; the text does not establish that the observed spreads persist out of sample or after implementation costs. The later-period results differ from the full-sample results, so factor effectiveness should not be treated as fixed across market regimes.

Key ideas

• The study constructs 56 trading and financial statement factors for Chinese A-share stocks.
• It tests factors by sorting stocks into ten groups and comparing the next month’s value-weighted returns of extreme groups.
• The report identifies 13 effective factors over 1997–2017 and 10 over 2007–2017.
• Many reported effective factors relate to trading frictions and liquidity.
• The authors assess factor correlations and note that orthogonalization can depend on factor ordering.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.