部分线性单指标模型在股票价格预测中的应用

发布时间：2018-08-20 14:06

【摘要】：部分线性单指标模型是由线性模型和单指标模型组合成的一类半参数统计模型。该模型能够在参数统计推断与非参统计推断之间取得某种平衡，具有良好的统计性质并且在经济、生物、医学等领域都有广泛的应用。随着我国经济和投资市场的不断发展，股票投资也受到人们的极大关注，且有越来越多人参与到股票市场中。股市在一个国家的金融领域扮演着至关重要的角色，而股票价格的波动是投资者最为关心的事情。由于股票交易市场存在着许多不定因素，这给投资者合理投资选股造成了一定困难，因此人们渴望能够对股价进行科学分析与预测。为此相关领域的许多学者作了大量的探索工作，并取得了一些行之有效的办法。基于部分线性单指标模型的灵活性，针对股价预测这一热点研究问题，本文考虑从上市公司公布的财务数据出发，利用部分线性单指标模型预测股价与财务指标之间的关系，进而对未来股价进行预测。本文主要完成以下几方面工作：第一，本文介绍了股票的相关概念及知识，包括股票的主要特征，股价的影响因素，影响股价的财务指标等。其中，对财务指标的介绍有助于选择合适的财务变量进行模型应用及股价预测。在此基础上，介绍了一些现有的关于股票价格预测的方法，并对每种方法的特点、优势、不足等进行了描述。第二，阐述了单指标模型及部分线性单指标模型的发展，概况，研究现状等，此外给出了模型中的未知参数和未知函数的估计值。第三，，实证分析。首先对所选财务指标进行降维处理，简化模型。然后应用部分线性单指标模型预测出降维后财务指标与股票价格之间的函数关系，并且与线性模型的预测结果进行对比，发现部分线性单指标模型的预测结果优于线性模型，因此本文提出的方法具有一定的应用价值。
[Abstract]:Partial linear single-parameter model is a kind of semi-parametric statistical model which is composed of linear model and single-parameter model. The model can achieve a certain balance between parametric statistical inference and non-parametric statistical inference. It has good statistical properties and is widely used in economic, biological, medical and other fields. With the development of our country's economy and investment market, people pay more and more attention to the stock investment, and more people participate in the stock market. The stock market plays a vital role in the financial sector of a country, and the volatility of stock prices is a matter of most concern to investors. There are many uncertain factors in the stock market, which makes it difficult for investors to make reasonable investment in stock selection, so people are eager to make scientific analysis and prediction of stock price. For this reason, many scholars in related fields have done a lot of exploration and made some effective methods. Based on the flexibility of partial linear single index model and aiming at the hot research problem of stock price forecasting, this paper considers the relationship between stock price and financial index by using partial linear single index model from the financial data published by listed companies. Then predict the future stock price. The main work of this paper is as follows: first, this paper introduces the related concepts and knowledge of stock, including the main characteristics of stock, the influencing factors of stock price, the financial index of stock price and so on. Among them, the introduction of financial indicators is helpful to select suitable financial variables for model application and stock price prediction. On this basis, some existing methods of stock price prediction are introduced, and the characteristics, advantages and disadvantages of each method are described. Secondly, the development, general situation and research status of single index model and partial linear single index model are described. In addition, the estimated values of unknown parameters and unknown functions in the model are given. Third, empirical analysis. First of all, the selected financial indicators are reduced to simplify the model. Then the partial linear single index model is used to predict the functional relationship between the financial index and the stock price after dimensionality reduction, and compared with the prediction result of the linear model, it is found that the prediction result of the partial linear single index model is better than that of the linear model. Therefore, the method proposed in this paper 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】：2013
【分类号】：F830.91;F224;O212.1

【参考文献】