基于支持向量机的股市预测

发布时间：2018-05-16 20:14

本文选题：支持向量机 + 技术分析　；参考：《北京工业大学》2005年硕士论文

【摘要】：股票市场是一个复杂的非线性动态系统,利用传统的时间序列预测技术很难揭示其内在的规律。本文,在分析传统预测分析方法的基础上,并考察了大量基于人工神经网络预测分析的方法基础上,采用支持向量机对上海股市进行预测。本文,首先对影响股市的因素进行分析,进而提出预测股市的参数选择策略,并对股市预测的过程进行了全面的介绍。本文,提出了一种可应用非线性支持向量机的参数选择方法,试验结果表明该方法是一种有效的方法。本文讨论了核函数及其参数对预测结果的影响。本文,采用标准支持向量机,对上证180 指数和上证综合指数,以及一些个股的走势和价格进行了预测,效果基本令人满意。然后,本文采用股市的实际数据,对支持向量机方法和BP 人工神将网络进行了对比。结果表明,支持向量机虽然在学习效果略逊于BP 网络,但在预测效果上优于BP 网络。
[Abstract]:Stock market is a complex nonlinear dynamic system. It is difficult to reveal its inherent law by using the traditional time series prediction technology. In this paper, based on the analysis of the traditional prediction and analysis methods, and a large number of artificial neural network forecasting methods, support vector machine (SVM) is used to forecast the Shanghai stock market. In this paper, the factors affecting the stock market are analyzed, and then the parameter selection strategy for forecasting the stock market is put forward, and the process of the stock market prediction is introduced comprehensively. In this paper, a parameter selection method based on nonlinear support vector machine is proposed. The experimental results show that this method is an effective method. In this paper, the influence of kernel function and its parameters on the prediction results is discussed. In this paper, the standard support vector machine is used to predict the trend and price of Shanghai Stock Exchange 180 Index and Shanghai Composite Index, as well as some individual stocks. The results are basically satisfactory. Then, using the actual data of stock market, the support vector machine method and BP artificial neural network are compared. The results show that support vector machine (SVM) is better than BP neural network in prediction, although its learning effect is inferior to BP neural network.
【学位授予单位】：北京工业大学
【学位级别】：硕士
【学位授予年份】：2005
【分类号】：F224.3

【引证文献】