股票价格预测的建模与仿真研究

摘要

研究股票价格准确预测问题,由于股票价格数据具非线性、随机性等变化规律,同时股票市场与国内外经济政治变化有关,传统股票价格预测方法只能对其线性变化规律进行准确预测,无法反映股票价格非线性部分进行有效建模,导致股价预测精度不高.为了提高股票价格预测精度,提出了一种遗传优化BP神经网络的股票价格预测模型.充分利用BP神经网络良好的非线性映射能力,对股票价格变化规律进行建模,并通过遗传算法对BP神经网络模型参数进行优化,从而获最优股票价格最优预测模型.实验结果表明,相对于传统股票价格预测模型,遗传算法优化BP神经网络的股票价格预测模型拟合程度更好,预测精度更高,为股票价格预测提供了依据.%Stock price data have the characteristics of nonlinear, randomicity, and etc, and tradition stock price prediction method is based on linear model, unable to accurately create the modele of the stock price, which causes the low prediction accuracy. In order to improve the prediction accuracy of stock prices, this paper put forward a method which used genetic algorithm to optimize the BP neural network of stock price prediction model. Making full use of the good nonlinear mapping capability of BP neural network, the stock price changing rules were modeled by using genetic algorithm in BP neural network training for global search capability. Experimental results show that com pared with the models of traditional BP neural network, the proposed model has better fitting degree, and the forecas ting accuracy is higher.