Publication of earnings forecasts developed by financial analysts is a new phenomenon in the Asian region. This study examines the bias and accuracy of analysts' earnings forecasts for firms in nine Asian countries and compares analyst forecasts to the random walk model forecasts. This study also performs regression analyses to identify the factors that explain differences in forecast accuracy. The results indicate the substantial inter-country variability in measures of forecasting bias and accuracy and the superiority of analysts' forecasts over the random walk model predictions. Analysts' forecasts for earnings per share are generally biased towards overestimation. The regression results identify some macroeconomic and firm specific factors that explain the differences in forecast accuracy across countries. Further analyses of forecast errors are also conducted with the sample segmented by two classifications, respectively.