Investor Learning About Analyst Predictive Ability

Abstract

Bayesian learning implies decreasing weights on prior beliefs and increasing weights on track records, as the latter become more precise. We test whether investors learn about analyst predictive ability in this manner by examining market reactions to analysts' forecasts. Consistent with investors shifting weight to track records, we find that market reactions to analysts' forecasts of quarterly earnings are increasing in the product of the accuracy and length of analysts' track records. Moreover, we show that the dynamic learning predicted by our model is more descriptive of market reactions to analysts' forecasts than a static model which predicts that investors' responses to forecast revisions condition only on the prior accuracy of the analyst.

Keywords: Analyst, Bayesian Learning

JEL Classification: G14, G24