针对股价指数特有的波动性,提出了基于灰色残差模型和BP神经网络的股指动态预测方法,并运用多元线性回归模型对两种动态预测结果进行拟合.同时,随机抽取部分上证指数和道琼斯指数的实证研究表明:动态预测模型能及时调整新数据对后续预测的影响,获得了较高的预测精度. In view of the unique volatility of the stock index, a stock index dynamic forecasting method based on gray residual model and BP neural network is proposed, and the multiple linear regression models are used to fit the two dynamic forecasting results.At the same time, some SSE indices and Dow Jones The empirical research shows that the dynamic prediction model can timely adjust the impact of the new data on the subsequent prediction and obtain a higher prediction accuracy.