【作者】 廖勇；

【导师】 唐常杰；

【作者基本信息】 四川大学 ， 计算机应用， 2005， 硕士

【摘要】 数据挖掘是从大量的数据信息中提取出隐含的知识、规律和行为模式的处理过程。经济学家一直致力于研究股票市场价格的变化,希望能从中找出一些规律,避免诸如股灾这种大的股市波动,从而保持经济稳定。股票市场是一个复杂的非线性系统,同时受多种因素的交互影响,对于股票未来价格的精确预测是非常困难的。股市预测被认为是当前时间序列预测中最富挑战性的应用之一,受到数据挖掘界的广泛关注[1]。基因表达式编程(GEP)是在遗传算法的基础上发展而来的遗传算法的新分支,它在个体的表示、个体的处理和结果的形式等方面与传统遗传算法有着显著的区别和优势。本文针对股票对象的特点,研究了用GEP 算法对股票指数和价格进行预测,取得了满意的效果。本文主要工作如下: 1. 阐述了数据挖掘的基本概念和流程; 2.对遗传类算法特别是GEP 算法的结构和特点进行了分析; 3. 对股票市场及数据的特点进行了研究和分析; 4.建立了基于基因表达式编程的股票分析模型GEP-STOCK 及其算法; 5.在GEP-STOCK 模型和算法的基础上对股票指数和价格进行了一系列实验,然后用神经网络算法进行了相应的对比实验,对实验结果进行了分析,实验结果表明,GEP 算法对股票指数预测的精度比神经网络算法高200%.更多还原

【Abstract】 Data mining is a process that discovers knowledge from mass data and information. It plays an important role in decision making and actions as guidance. Gene Expression Programming (GEP) is a new member in the family of Genetic Algorithm (GA). It is different from traditional GA in expressing and processing of individual and form of result. The economists worked hard to research the change of the stock-market,hope to find some rules to avoid the big waving of stock-market such as stock-disaster and keep the stability of the stock-market.The stock-market is a complicated unlinear system infected by many factors in the same time,the accurate prediction of the stock-price is very difficult,the prediction of the stock-market is regarded as one of the most chanlleging applications in the time-series predicting,,it attracts far attention in the Data-Mining area. Based on the features of stock objects, this paper have researched the method to predicting the stock-index and the stock-price by GEP algorithm and got satisfying result. The main contrition of this work includes: 1. Foamally describing the basic concept and diagram of Data-Mining; 2. Analysing the features of genetic algorithm especially the GEP algorithm; 3. Researching and analyzing the features of stock-market and stock-data; 4. Building the the stock analyse model named GEP-stock based on GEP , and giving its algorithms; 5. Giving experiments for stock-index and stock-price based on the GEP-Stock model,.anylysing the experiment results,compare it with neural networks. The experiments show that the precision of new model is much higher than trandisiona method in neral networks. The thises is organized as follows: Chapter 1 intoduces the basic concepts of Data-Mining.Chapter 2 analyses the features of stock-data.Chapter 3 analyses the features of Genetic-algorithm.Chapter 4 introduces the GEP algorithm.Chapter 5 presents the GEP-STOCK model including the STOCK-GENE and the STOCK-fitness that appropriated to the special rules of stocks. Chapter 6 verifies the effectiveness of STOCK-GEP algorithm by experiments and analyses on the real stock-price index and stock-price.Chapter 7 gives conclusion and discusses prospects for the future works.更多还原