部分线性模型在股票价格预测中的应用研究

发布时间：2018-01-03 00:40

本文关键词：部分线性模型在股票价格预测中的应用研究　出处：《辽宁师范大学》2012年硕士论文　论文类型：学位论文

【摘要】：随着我国经济的快速发展和完善，证券投资特别是股票交易正逐渐发展成为当今经济社会的重要环节。股票是市场经济的产物，股票的发行与交易促进着市场经济的发展。股票预测分析是指以准确的统计资料和股市信息为依据，从股市的历史、现状和规律出发，运用科学的方法对股票未来发展状况进行测定。从我国目前股票交易市场情况出来看，，中国股票交易市场存在着很多的不确定因素。同时股票预测还是理性投资、科学投资的重要前提。根据这些具体情况，本文主要工作如下：第一，本文介绍了股票的相关基础知识，特别介绍了影响股票价格的公司主要财务指标。在这些基础知识之上，介绍了已有的股票价格预测方法，并分析了每种方法的优缺点。第二，阐述了本文所应用于股票预测的统计模型。本文的统计模型，主要包括部分线性模型和主成分分析两部分。首先对部分线性模型进行概述，并且讨论了部分线性模型的最小二乘估计，最后介绍了主成分分析原理。第三，本文通过中国银河证券海王星股票分析软件，得到传统能源板块的66支股票，去除3支ST股票和数据缺失、异常的11支股票，剩余的52支股票的2010年12月30日公布的公司财务指标和个只股票的收盘价作为回归分析的样本数据。先运用相关性分析、单元回归等方法对个只股票的公司财务指标进行筛选筛选，其次利用主成分分析方法对筛选出的财务指标数据进行降维处理，最后基于降维后的数据拟合部分线性模型。并且根据2011年9月30日公布的财务指标，应用本文统计模型对当日股票收盘价进行预测，通过与实际股票收盘价相比较。同时还应用线性模型对股票价格进行预测，将两种方法的预测结果相对比说明本文方法具有一定的应用价值。
[Abstract]:With the rapid development and improvement of China's economy, securities investment, especially stock trading, is gradually developing into an important part of the economic society. Stock is the product of market economy. Stock issue and trading promote the development of market economy. Stock prediction and analysis is based on accurate statistical data and stock market information, starting from the history, current situation and law of stock market. Using scientific methods to determine the future development of stocks. From the current stock market situation in China to see. There are many uncertain factors in China's stock market. At the same time, stock forecasting is also an important prerequisite for rational investment and scientific investment. According to these specific conditions, the main work of this paper is as follows: First, this paper introduces the basic knowledge of the stock, especially the main financial indicators that affect the stock price. On the basis of these knowledge, the paper introduces the existing methods of stock price prediction. The advantages and disadvantages of each method are analyzed. Secondly, this paper describes the statistical model used in stock forecasting. The statistical model of this paper mainly includes two parts: partial linear model and principal component analysis. Firstly, the partial linear model is summarized. The least square estimation of partial linear model is discussed, and the principle of principal component analysis is introduced. Thirdly, through the software of Neptune stock analysis of China Galaxy Securities, this paper obtains 66 stocks of traditional energy plate, removes 3 St stocks and missing data, and 11 abnormal stocks. The financial indexes of the remaining 52 stocks published on December 30th 2010 and the closing price of each stock were used as the sample data of regression analysis. First, the correlation analysis was used. The unit regression method is used to screen the financial index of a stock, and the principal component analysis method is used to reduce the dimension of the selected financial index data. Finally, based on the reduced dimension data fitting partial linear model, and according to the financial indicators published on September 30th 2011, this paper uses the statistical model to forecast the closing price of the stock on that day. Compared with the actual stock closing price, the linear model is also used to predict the stock price. The comparison of the results of the two methods shows that this method has certain application value.
【学位授予单位】：辽宁师范大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F224;F832.51

【参考文献】