HARRV, HARRVJ and HARRVJARCH are proposed for the heterogeneity in Chinese stock market, and the least square method is used to combine the Newey West covariance to estimate the parameters for forecasting and then to compare the fitting effect and forecasting effect of the three models. The empirical results show that the heterogeneity of the return volatility of the stock market is mainly from monthly fluctuation (longterm trader) and jump fluctuation but daily fluctuation (shortterm trader) and weekly fluctuation (middleterm trader) have small effect, and that HARRVJARCH model has better fitting effect. The results indicate that HARRVJARCH model can better describe the heterogeneity in stock market.