| This study examines, both analytically and empirically, how the properties of interim earnings affect analysts' annual earnings forecast revisions around interim earnings announcements. My rational expectations model of analysts' forecast revisions allows analysts to acquire private information in both the pre-announcement and announcement periods. In each period, the properties of the private information are determined by the property of the public information it will be used to interpret. The model predicts the following: (1) Analysts' average forecast revisions are positively associated with both the surprise and precision of interim earnings. The associations are stronger the earlier the announcements are in the year. (2) The change in analysts' forecast dispersion is positively associated with the interim earnings surprise and pre-announcement information precision and negatively associated with interim earnings precision. (3) Analysts' forecast jumbling is positively associated with the surprise and precision of interim earnings and negatively associated with pre-announcement information precision. Using I/B/E/S analysts' earnings forecast data and the pooled OLS linear regression method, I find the empirical evidence that is consistent with the predictions about analysts' average forecast revisions and the change in forecast dispersion. However, the empirical evidence does not reject the null hypothesis about analysts' forecast jumbling. |
