Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We evaluate the performance of various methods for estimating factor returns in an approximate factor model. Differences across estimators are most pronounced when there is cross-sectional heteroscedasticity or when cross-sectional sample sizes, n, have fewer than 4,000 assets. Estimators incorporating either cross-sectional or time-series heteroscedasticity outperform the other estimators when those types of heteroscedasticity are present. The differences are most pronounced when the cross-sectional sample is small.Received December 2, 2015; editorial decision May 16, 2017 by Editor Jeffrey Pontiff.