摘要
本文给出了一种基于BP神经网络的股票市场建模、预测以及决策方法。应用神经网络进行股票中期预测,输入数据的复杂性给网络训练效率和预测精度造成了显著的负面影响。我们应用模糊曲线分析法进行了输入变量的筛选,该方法主要是用来压缩输入数据的维度,发现影响产出变量的重要因素。它通过求相关度,贡献弹性,根据样本点拟合样本曲线,最后选取出影响变量的重要因素。结果表明,经该方法处理后的数据输入神经网络不仅减少了输入数据量,使训练时间减少,运算速度提高,而且预测精度有了明显的改善。
This paper presents a method based on the BP neural network for stock market modeling, forecasting and decision-making. In NN- based stock forecasting, the complexity of input data has a negative effect on network training efficiency and forecasting precision, Focusing on solving this problem, a fuzzy curve method is developed to filter variables. This method is developed to eliminate those inputs, which are dependent on other important inputs. The process is fast and can be effectively used on the systems with a large number of inputs and data points. The results show that after data processing, the training time is reduced and the forecasting precision is enhanced.
出处
《计算机工程与科学》
CSCD
2006年第5期115-117,共3页
Computer Engineering & Science
关键词
模糊曲线法
神经网络
股票预测
fuzzy curve method
neural network
stock forecasting
