Upper-Tail Anomaly Correlation as an Equity Selection Factor

Summary

The research note studies tail dependence between individual stocks and the broader market. Using copulas, it estimates upper-tail and lower-tail correlation, which describe whether assets tend to move together during extreme outcomes. The reported standalone tail-correlation factors showed stock-selection associations in the China All Share universe, but the authors found they were highly related to idiosyncratic volatility; removing that exposure eliminated their selection effect.

The authors then combine the two tail measures into an upper-tail anomaly correlation factor and report positive rank information coefficients across the China All Share, CSI 500, and CSI 300 universes over a historical sample from 2006 through September 2018. They also report comparatively low rank-IC correlations with other factors after neutralization, suggesting distinct information. These are historical factor results, not evidence of future returns; the note cautions that extreme markets may disrupt results and that models can lose effectiveness.

Key ideas

• Copula methods are used to estimate stock-market dependence in upper and lower tails.
• The initial tail-correlation factors were strongly associated with idiosyncratic volatility.
• An upper-tail anomaly factor showed stock-selection associations in three Chinese equity universes.
• After style neutralization, the factor showed comparatively low rank-IC correlations with other factors.
• The historical findings may not persist, especially in extreme market conditions.

Tags

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