Analysis and comparison of financial analysts', time series, and combined forecasts of annual earnings

Abstract

This paper examines the effects of disagreement in financial analysts' earnings forecasts on the accuracy of analysts', time series, and combined forecast made at four forecast horizons. The empirical analysis indicates that, while analysts do better than any of the three time series models studied, simple combinations of analysts' and time series forecast are superior to forecast from either source at every horizon. The accuracy of individual and combined forecasts is inversely related to the dispersion of analysts' forecast, and the improvements from combination is greatest when analysts' forecast dispersion is greatest and the forecast horizon longest.