Are revisions of security analysts' earnings forecasts rational? Canadian evidence

Research conducted considers the quality of security analysts' earnings forecasts by testing the rationality of forecast revisions for Canadian companies. It appears that when making small positive revisions, analysts overreact; in subsequent periods, forecasts are revised downward. Large positive revisions are followed by significant positive revisions, suggesting underreaction by analysts. Overall, for positive revisions, while both modest over- and underreaction is evident, neither tendency is pronounced. For negative revisions, regardless of their size, significant underreaction is suggested; when initially forecasting earnings decreases, analysts appear to underestimate the decline in earnings. This is consistent with the view that forecasts are optimistic.