ACCURACY IMPROVEMENTS FROM A CONSENSUS OF UPDATED INDIVIDUAL ANALYST EARNINGS FORECASTS

This study compares the bias and accuracy of mean consensus earnings forecasts comprised of updated individual analyst forecasts to the bias and accuracy of other frequently cited forecast measures. It is well known that analyst forecast accuracy improves as the forecast horizon shortens. I use a model that updates individual analyst forecasts for information revealed after the forecast date, in effect shortening the forecast horizon. Any resulting accuracy improvement is of interest to investors, who rely on analysts for earnings forecasts, and is relevant to studies that compare the accuracy of analyst forecasts with time series-based forecasts and management forecasts. My study adds to the literature on combining forecasts by demonstrating that a mean consensus of updated, chronologically synchronized, individual analyst forecasts is a less biased and more accurate forecast of actual earnings than other frequently cited forecast measures, including an average of recently issued forecasts.