An Evaluation of Security Analysts' Forecasts.

Abstract ABSTRACT: Recent literature in accounting, finance, and economics often assumes that information can be processed efficiently. Among the outputs of the processing activity are the presumably appropriate assessments of the underlying probability distributions for all important variables, and a good deal of the recent research assumes that observable realizations of the variables are drawn from these distributions. This paper provides evidence concerning the ability of selected individuals, namely security analysts, to provide estimates of earnings per share after presumably processing the available information. Several aspects of the quality of analyst forecasts are examined. The study indicated, as expected, that analysts' forecasts become more accurate as the reporting date is approached. Furthermore, the predictions of changes in earnings per share data contain no significant systematic bias. However, the authors do not find sufficient support for the expected decline in forecast variability among analysts as the reporting date is approached.