About Chinese Brokerage Analysts Forecast Error Influence Factors Of The Empirical Research

In this paperⅠinvestigate the data of individual analyst forecasts in China and find the overoptimistic bias exists in forecasting. But the comparative performance of analyst forecasts is still better than naive model. The influential factors of analysts forecast errors are further discussed through two models, time serial model and cross sectional model. Previous researches suggest serial correlation exists in current forecast errors and previous ones. While cross sectional model suggests that the forecast errors are due to cross sectional variables such as time factors, company factors, analyst factors and environmental factors. In time serial model, earning surprise is found to be related to forecast errors. For cross sectional model, it is found that the forecasts are more conservative and accurate before financial crisis. It is also found that analysts tend to be more optimistic and less accurate for bigger companies. Among analyst factors, the rank of analyst, which represents his or her experience and ability, helps to improve the accuracy of forecasting. Besides, both macroeconomic factors such as total output and inflation rate and financial market factors such as stock indexes have significant influence on analysts forecast errors.In the final part, both factors suggested by time serial model and cross sectional model are included to build up a model with higher explanatory power. This model suggests that the environmental factors such as total output, inflation and general condition of stock market (stock index) and earning surprise have the most significant causality to analysts forecast errors.