Arbitrage risk and the cross-section of stock returns: Evidence from China

Yu En Lin, Chien Chi Chu, Akihiro Omura, Bin Li, Eduardo Roca

Research output: Contribution to journalArticle

2 Citations (Scopus)


We demonstrate that arbitrage risk, constructed using three measures — noise trader risk, trading cost and information uncertainty — can predict the return of stocks cross-sectionally in China. The findings are broadly consistent even when out-of-sample tests are conducted using the Fama-MacBeth cross-sectional regression approach. We also construct hypothetical portfolios using the information arising from arbitrage risk and find the existence of abnormal returns which is robust to the use of various portfolios constructed by re-sampling the observations through multiple approaches (e.g., by market capitalization and by book-to-market ratio). Lastly, we reconstruct our portfolios by considering the unique nature of the Chinese stock market (e.g., the dominance of individual investors). Our trading strategies again successfully obtain abnormal returns, suggesting that arbitrage risk can be useful to construct effective investment portfolios in China.

Original languageEnglish
Article number100609
JournalEmerging Markets Review
Publication statusPublished - Jun 2020


  • Arbitrage cost
  • Arbitrage risk
  • Information uncertainty
  • Out-of-sample forecast returns
  • Transaction cost

Fingerprint Dive into the research topics of 'Arbitrage risk and the cross-section of stock returns: Evidence from China'. Together they form a unique fingerprint.

  • Cite this