基于遗传算法的小波神经网络在资本市场中的应用

发布时间：2018-01-16 00:34

本文关键词：基于遗传算法的小波神经网络在资本市场中的应用　出处：《复旦大学》2013年硕士论文　论文类型：学位论文

【摘要】：最近二十年,以上海和深圳两大证券交易所的开幕为契机,中国的金融市场发生了翻天覆地的变化。金融相关行业,银行,证券,保险分别都有了一套完整的监管体系,行业人才也在日益增加,证券分析师们也都在寻找着在证券市场更加行之有效的证券分析方法。数学作为金融学研究中一种必不可少的工具,正在证券分析的领域发挥越来越大的作用。而人工神经网络也在其中发挥了重要的作用。基于反向传播算法的小波神经网络有着很好的学习能力,良好的容错性,并且可以并行处理信息,通过调整各节点之间的权值,连接关系,可以达到很好的处理信息的作用。遗传算法以及量子遗传算法在函数的逼近以及网络权值的运算中可以得到良好的效果。本文以中国股票市场中的板块以及个股作为研究对象,使用基于遗传算法和基于量子遗传算法的小波神经网络,并结合反向传播算法,力图寻找证券板块的每日趋势以及个股走向中的规律。使用小波神经网络,我们可以输入历史的数据,得到预测的数据。并与历史已有数据相对比,并最终得到了比较好的模拟结果,验证了神经网络的有效性。
[Abstract]:In the last two decades, with the opening of two major stock exchanges in Shanghai and Shenzhen, China's financial market has undergone earth-shaking changes. Insurance has a complete regulatory system, industry talent is increasing, and securities analysts are looking for more effective securities analysis methods in the stock market. Mathematics is an indispensable tool in financial research. It is playing a more and more important role in the field of securities analysis, and artificial neural network also plays an important role. Wavelet neural network based on backpropagation algorithm has a good learning ability. Good fault tolerance, and can process information in parallel, by adjusting the weight of each node, join the relationship. Genetic algorithm and quantum genetic algorithm can get good results in function approximation and network weight operation. In this paper, we use the wavelet neural network based on genetic algorithm and quantum genetic algorithm, and combine the back propagation algorithm with the plate and individual stock in Chinese stock market as the research object. Using wavelet neural network, we can input historical data, get predicted data, and compare with historical data. Finally, good simulation results are obtained to verify the effectiveness of the neural network.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP18;F832.51

【相似文献】