Evidence on the Possible Underweighting of Earnings-Related Information
研究分析师和投资者是否系统性地低估了盈利预测误差的持续性，并探讨投资者是否在分析师修正预测时重新评估这种持续性，为理解盈利公告后漂移现象提供新证据。
This study proposes and tests hypotheses related to three issues. (1) Do analysts underestimate the persistence (permanent component) of earnings forecast errors? (2) Do investors use analysts' earnings forecast revisions to reassess the persistence of previous forecast errors? (3) Do investors systematically underestimate the persistence of previous forecast errors signaled by forecast revisions? In their investigations of post-earnings-announcement drift, Rendleman, Jones, and Latan6 [1987], Bernard and Thomas [1989; 1990], and Freeman and Tse [1989] present evidence suggesting that investors systematically underestimate the persistence of earnings forecast errors. I use analysts' expectations (their forecasts) to test the hypothesis that market participants underestimate persistence (issue (1)). Freeman and Tse also show that investors reevaluate the persistence of earnings innovations in the light of subsequent earnings announcements. I investigate whether investors similarly reassess persistence in the light of subsequent analyst earnings forecast revisions (issue (2)). Finally, if investors both underestimate the persistence of forecast errors at earnings announcements and reevaluate persistence when forecasts are revised, then they may underestimate the persistence of earnings forecast errors signaled by forecast revisions. That is, investors may