EMD和GARCH模型应用于股票价格预测

将EMD（经验模式分解）方法应用到股票价格趋势的预测中，找出影响股票市场波动的关键因素，旨在提高预测的精确性。通过EMD方法将上证指数日收盘价数据分解为不同频率的数据段，重组为高频序列、低频序列和趋势序列，运用高阶自回归和GARCH模型对分解出来的各序列进行拟合和预测，避免各个分段预测过程中的误差累积，最后对预测数据重组，得到样本外数据的预测序列。结果表明，该模型具有较好的预测效果，能给投资者提供更为合理的股票投资意见，同时为趋势预测研究提供借鉴。

Stock Price Forecasting Based on EMD and GARCH Model

The EMD（Empirical Mode Decomposition） method was applied to stock price trend forecast,for improving the forecasting accuracy. Using the EMD method we decomposed the daily closing price data of Shanghai Stock Index into the data segments with different frequency,then fitted and forecasted the data segments through the high order autoregressive and the GARCH model. At last,we get the out-of-sample forecast sequence after the restructure of the prediction data. The results show that the model has good prediction effect,can provide more reasonable stock investment advice to investors,and has reference value for trend prediction research.