And the Cross-Section of Expected Returns

Harvey, Liu and Zhu’s 2016 Review of Financial Studies paper (NBER WP 20592, 2014) censuses the “factor zoo” — several hundred factors claimed in top journals to explain the cross-section of equity returns — and argues that, given this extensive collective data mining, the conventional single-test hurdle of a t-statistic above 2.0 is statistically meaningless. It introduces a multiple-testing framework adapting Bonferroni, Holm and Benjamini-Hochberg-Yekutieli corrections to correlated and incomplete tests, and concludes a newly discovered factor should clear a t-ratio above roughly 3.0. It appears in this vault as the canonical reference for Data-Snooping Bias across a literature.

Connections

• Data-Snooping Bias — defines, source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2249314
• Overfitting in Quantitative Trading — relates, source: https://www.nber.org/papers/w20592
• Campbell Harvey — proposes_model, source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2249314
• Yan Liu — proposes_model, source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2249314