This research uses four different measures of bold to investigate how prior information affects analyst herding decisions. Results for the more restrictive measures of bold suggest that the probability of herding is greater with large information shocks. Evidence also shows that analysts are more likely to herd in their forecast revisions when their current outstanding forecasts deviate more from the consensus mean and in the presence of strong observable signals. In general, analysts with current outstanding forecasts that are optimistic are more likely to issue revised forecasts that are also optimistic.
목차
Abstract I. Introduction II. Prior Literature and Hypothesis Development III. Sample Selection IV. Research Design and Model Development 4.1. Measurement of Bold (Herding) Forecasts 4.2. Model Development V. Results 5.1. Descriptive Statistics 5.2. The Association between Forecast Boldness and Prior Information 5.3. The Effect of Analyst Optimism on Forecast Revision Boldness 5.4. The Effect of Signal Strength on Bold Forecasts 5.5. Controlling for Analyst Characteristics VI. Summary and Conclusion References Table