Title: A model of persistent analyst biases to improve the earnings consensus forecast
Abstract:This paper explores to what extent the assumption, that analysts have a bias that is persistent over time, can be employed to improve the consensus forecast. We estimate the analyst’s individual biase...This paper explores to what extent the assumption, that analysts have a bias that is persistent over time, can be employed to improve the consensus forecast. We estimate the analyst’s individual biases and establish that our model yields on average a mean forecast error that is 40 % smaller than the consensus forecast and a root mean-squared error that is 2 %below the consensus forecast. The reductions in mean-squared error range from an out-performance of 29 % to an under-performance of 2 % in our sample period spanning fiscal years 1993 to 2008. Regressions of earnings per share on the two earnings forecasts show that our model forecast helps improve on the consensus forecast. We also analyze the eect of earnings surprises on abnormal returns and find that our model captures the information content of analysts’ forecasts better than the consensus forecast. In particular, depending on whether we estimate within the sample or out-ofsample, the earnings response coecientRead More
Publication Year: 2014
Publication Date: 2014-01-01
Language: en
Type: article
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot