论文部分内容阅读
针对股票市场中价格序列是一个复杂的非线性动态系统,同时难以实现准确预测的问题,采用RBF神经网络方法,利用其较强的非线性处理能力进行股票价格预测研究,同时利用具有全局搜索能力的遗传算法对RBF神经网络进行优化研究,得到性能更加优越的神经网络模型.分别使用传统RBF神经网络和遗传算法优化后的RBF神经网络进行股票价格预测,实验结果表明:利用遗传算法优化后的RBF神经网络在网络的结构和逼近性能上都有明显改进和提高,能够有效地反映股票价格的波动特性,提高股价预测的准确性.该研究成果对股票市场规律的研究具有一定的参考价值和指导意义.
Aiming at the problem that the price series in the stock market is a complex nonlinear dynamic system and it is difficult to predict accurately, RBF neural network method is used to forecast the stock price with its strong non-linear processing ability. At the same time, Genetic algorithm to optimize the RBF neural network to get a more superior performance neural network model.The RBF neural network and the genetic algorithm optimized RBF neural network were used to predict the stock price.The experimental results show that the genetic algorithm RBF neural network has obvious improvement and improvement in network structure and approximation performance, which can effectively reflect the fluctuation characteristics of stock price and improve the accuracy of stock price prediction.The research results have certain reference value to the study of stock market law and Guiding significance.