Power Sorting

We propose a novel approach for constructing characteristic-based equity factors, termed “power sorting”. Power sorting exploits the non-linearities and asymmetries inherent in the characteristic-return relations, while it remains computationally simple and avoids excessive weights. We demonstrate that power sorting achieves consistently superior out-of-sample performance compared to traditional quantile sorting and other factor portfolio construction methods. Our results are pervasive across factors, robust through time, cannot be attributed to increased turnover or tail risk, and extend to multi-factor strategies. Finally, we show that power sorted versions of well-known asset pricing factor models outperform the original ones.

Read full research paper here.