View article

This study uses experiments to examine whether individuals' earnings forecasts correctly reflect the time series properties of quarterly earnings, in particular, the positive autocorrelation in seasonal quarterly changes and the negative fourth-order moving average term documented by Brown and Rozeff (1979). We find that individuals' forecasts are sensitive to the magnitude of these time series components; however, individuals typically underweight the moving average term and under- (over-)weight the most recent seasonal quarterly change when it has a strong (weak) effect on future earnings. Individuals also place slightly more weight on quarterly changes when earnings are reported relative to those four quarters prior. These results suggest that the documented stock market under-reaction to quarterly earnings may not hold universally; rather, it may be composed of under-reactions to firms with strong …