基于组合预测模型的股票价格趋势预测

Application of Combination Forecasting Model in Stock Price Prediction

针对股票市场高度非线性的特点,单一预测模型很难描述出股票价格趋势的整体特征,提出了一种金融时序预测的组合预测模型。首先,利用自回归移动模型(ARIMA)对股票价格线性趋势进行预测。然后,利用回归支持向量机(SVR)模型对非线性随机变化规律进行预测。最后,采用模糊时变权重方式对两种模型进行结合,得到一种综合考虑股票价格线性和非线性的预测模型。仿真结果显示,组合预测模型取得了令人满意的效果。

For highly nonlinear characteristics of the stock market and a single forecasting model difficult to describe the overal characteristics of the stock price trends,this paper proposes a combination forecasting model to predict.First,Autoregressive lntegrated Moving Average Model （ARlMA） is used for the stock price linear trend prediction and Support Vector Machine regression （SVR） model is used to predict nonlinear stochastic variation.Final y,this paper use the fuzzy variable weight way to combine the two models to obtain a model with linear and nonlinear consideration.