单指标模型在证券投资风险预测中的应用
发布时间：2018-07-26 19:33
【摘要】：单指标模型是一种广义的回归模型－半参数回归模型，该模型能够将多维变量降到一维指标，解决了多元非参数回归无法克服的维数问题，具有良好的统计性质并且在生物医学和金融经济等领域都有广泛的应用。 随着我国经济和证券市场的不断发展，，企业证券投资受到人们的极大关注，且有越来越多人参与到企业证券投资的市场中。企业证券投资在一个国家的金融领域扮演着至关重要的角色，而证券投资风险控制是投资者进行财务管理的一项重要内容。由于市场存在着许多不定因素，如何对证券投资风险进行预测、控制，这给投资者做出正确投资决策造成了一定困难，因此人们渴望能够对企业的现状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。 基于单指标模型的灵活性，针对企业证券投资这一热点研究问题，本文考虑从上市公司公布的财务数据出发，利用单指标模型预测企业证券投资风险与财务指标之间的关系，进而对未来投资风险进行预测，并与普遍被大家所接受的Logistic回归模型预测方法进行比较。本文工作安排如下： 第一，本文介绍了证券投资风险的相关概念及知识，包括风险的主要特征，证券投资风险的影响因素，影响证券投资风险的财务指标等。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证券风险预测。在此基础上，介绍了一些现有的关于证券投资风险预测的方法，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。然后应用单指标模型预测出降维后财务指标与证券投资风险之间的函数关系，并且与Logistic回归模型的预测结果进行对比，发现单指标模型的预测结果优于Logistic回归模型，因此本文提出的方法具有一定的应用价值。
[Abstract]:The single index model is a generalized regression model-semi-parametric regression model. The model can reduce the multidimensional variables to one dimension index, and solve the dimension problem that can not be overcome by multivariate non-parametric regression. It has good statistical properties and is widely used in biomedical, financial and economic fields. With the development of our country's economy and securities market, people pay more and more attention to the enterprise's securities investment, and more people participate in the market of enterprise's securities investment. Corporate securities investment plays an important role in the financial field of a country, and the risk control of securities investment is an important part of the financial management of investors. Because there are many uncertain factors in the market, how to predict and control the risk of securities investment makes it difficult for investors to make the right investment decision, so people are eager to be able to scientifically analyze and predict the present situation of enterprises. For this reason, many scholars in related fields have done a lot of exploration and made some effective methods. Based on the flexibility of single index model and the hot research problem of enterprise securities investment, this paper considers the relationship between securities investment risk and financial index by using single index model from the financial data published by listed companies. Then the future investment risk is forecasted and compared with the commonly accepted Logistic regression model. The work of this paper is as follows: first, this paper introduces the related concepts and knowledge of securities investment risk, including the main characteristics of the risk, the influence factors of the securities investment risk, the financial index that affects the securities investment risk, and so on. Among them, the introduction of financial indicators is helpful to select suitable financial variables for model application and securities risk prediction. On this basis, some existing methods of risk prediction of securities investment 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 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 single index model is used to predict the functional relationship between the financial index and the risk of securities investment after dimensionality reduction, and compared with the prediction results of the Logistic regression model, it is found that the prediction result of the single index model is better than that of the Logistic regression 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】：2014
【分类号】：F224;F830.91
本文编号：2147069
[Abstract]:The single index model is a generalized regression model-semi-parametric regression model. The model can reduce the multidimensional variables to one dimension index, and solve the dimension problem that can not be overcome by multivariate non-parametric regression. It has good statistical properties and is widely used in biomedical, financial and economic fields. With the development of our country's economy and securities market, people pay more and more attention to the enterprise's securities investment, and more people participate in the market of enterprise's securities investment. Corporate securities investment plays an important role in the financial field of a country, and the risk control of securities investment is an important part of the financial management of investors. Because there are many uncertain factors in the market, how to predict and control the risk of securities investment makes it difficult for investors to make the right investment decision, so people are eager to be able to scientifically analyze and predict the present situation of enterprises. For this reason, many scholars in related fields have done a lot of exploration and made some effective methods. Based on the flexibility of single index model and the hot research problem of enterprise securities investment, this paper considers the relationship between securities investment risk and financial index by using single index model from the financial data published by listed companies. Then the future investment risk is forecasted and compared with the commonly accepted Logistic regression model. The work of this paper is as follows: first, this paper introduces the related concepts and knowledge of securities investment risk, including the main characteristics of the risk, the influence factors of the securities investment risk, the financial index that affects the securities investment risk, and so on. Among them, the introduction of financial indicators is helpful to select suitable financial variables for model application and securities risk prediction. On this basis, some existing methods of risk prediction of securities investment 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 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 single index model is used to predict the functional relationship between the financial index and the risk of securities investment after dimensionality reduction, and compared with the prediction results of the Logistic regression model, it is found that the prediction result of the single index model is better than that of the Logistic regression 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】：2014
【分类号】：F224;F830.91
【参考文献】
相关期刊论文 前3条
1 邢有洪;;上市公司财务危机预警研究[J];财会通讯;2013年08期
2 吴世农,韦绍永;上海股市投资组合规模和风险关系的实证研究[J];经济研究;1998年04期
3 朱永忠;姚烨;张艳;;基于主成分分析和Logistic回归的上市公司财务困境预警模型的研究[J];浙江工业大学学报;2012年06期
本文编号：2147069
本文链接：https://www.wllwen.com/jingjilunwen/jinrongzhengquanlunwen/2147069.html
