Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The authors show that the returns to the typical long-term contrarian strategy implemented in previous studies are upwardly biased because they are calculated by cumulating single-per iod (monthly) returns over long intervals. The cumulation process not on ly cumulates "true" returns but also the upward bias in single-period reutrns induced by measurement errors. The authors also show that the remaining "true" returns to loser or winner firms have no relation to overreaction. This study has important implicati ons for event studies that use cumulative returns to assess the impact o f information events. Copyright 1993 by American Finance Association.