净资产收益率波动率及其与现金流的关系研究

发布时间：2018-06-20 02:38

本文选题：净资产收益率 + 波动率　；参考：《成都理工大学》2014年硕士论文

【摘要】：净资产收益率是一个综合性极强的财务指标，直接反应了一个公司的盈利能力，是现代成熟市场中广大投资者最为倚重的参考指标，在投资领域一直备受各界关注。本文借助已有研究成果，选取44家中国上市公司1994年--2013年净资产收益率数据为研究对象，对净资产收益率波动情况进行了实证研究，并分析了净资产收益率波动率与现金流相互之间的动态关系，为投资者、经营管理者以及后续研究提供实证经验和建议。本文主要研究内容有：（1）引言。该部分首先阐述了本文的研究背景、意义，包括问题的提出，并明确界定了相关概念，然后介绍了本文的研究思路及方法，最后是本文的创新点及不足。（2）文献综述。该部分主要是对国内外关于净资产收益率、波动率模型以及向量自回归模型的文献的研究，并归纳其主要观点和研究方法，为本文实证研究指明方向。（3）波动率模型概述。该部分详细介绍了各种广泛应用于金融领域的主流波动率模型，并通过分析比较，评选出最适合本文建模的类型。（4）净资产收益率波动率实证分析。该部分建立了多个波动率模型，并对其进行综合评价，找出相对较优的模型，用以净资产收益率波动率预测以及检验净资产收益率波动率中的不对称性影响。（5）净资产收益率波动率与现金流之间的动态关系的实证研究。该部分先是以净资产收益率波动率和现金流为变量建立向量自回归模型，并借助Granger因果关系分析法和脉冲响应函数法进行结构分析；再以净资产收益率和现金流为变量建立向量自回归模型，并借助方差分解进行结构分析。（6）结论。该部分对本文所做研究做出总结，并对存在的问题进行说明，最后指出未来研究方向。本文首先通过文献比较以及理论演绎，得出结论：GARCH模型对金融数据时间序列的拟合效果和预测能力较好，且预测期限越短预测能力越好；EGARCH模型在捕捉波动率中的非对称性影响中表现较好；已实现波动率对高频交易数据的预测能力较好。本文借助GARCH模型和EGARCH（1，1）模型，以及另外几种基本的GARCH扩展模型--GARCH-M模型、TGARCH模型和PGARCH模型来实证分析净资产收益率波动情况，结果表明：GARCH（1，1）模型对中国上市公司净资产收益率的拟合效果和预测能力相对较好，，EGARCH（1，1）模型证实了中国上市公司净资产收益率波动率中存在较为显著的非对称性影响。关于净资产收益率波动率与现金流之间的相互关系的实证研究。本文先是以净资产收益率波动率和现金流为变量建立VAR模型，并借助Granger因果关系分析法和脉冲响应函数分析法进行结构化分析，结论显示：净资产收益率波动率与现金流之间存在双向的Granger因果关系，即二者互为对方的Granger原因；净资产收益率波动率受自身残差冲击的影响大于受现金流残差的影响，且受自身影响时响应更为迅速，同时，现金流受自身残差冲击影响大于受净资产收益率波动率的残差影响。其次以净资产收益率和现金流为变量建立VAR模型，并利用方差分解进行结构化分析，结论显示：净资产收益率波动率变动由自身残差所导致的比例远大于由现金流残差所导致的比例，同样现金流方差变动由自身残差贡献的比例也更大，这种贡献比例在冲击产生时会发生适当的变化，并在未来某期保持稳定。
[Abstract]:Net asset yield is a comprehensive financial index, which directly reflects the profitability of a company. It is the most important reference index for the majority of investors in the modern mature market. In the field of investment, all walks of life have been paid attention to. In this paper, the net assets income of 44 Chinese listed companies in 1994 is selected with the help of the existing research results. The rate data is the research object, the fluctuation of net asset returns is empirically studied, and the dynamic relationship between the volatility of net assets yield and the cash flow is analyzed. The empirical experience and suggestions are provided for the investors, managers and follow-up research.
The main contents of this paper are as follows:
(1) introduction. This part first expounds the background and significance of this paper, including the proposal of the problem, and clearly defines the relevant concepts, and then introduces the research ideas and methods of this article, and finally is the innovation and deficiency of this article.
(2) literature review. This part mainly studies the literature on net asset returns, volatility model and vector autoregressive model, and sums up its main viewpoints and research methods, which indicates the direction of the empirical study.
(3) an overview of the volatility model. This section introduces a variety of mainstream volatility models widely used in the financial field, and selects the most suitable models for this model through analysis and comparison.
(4) an empirical analysis of the volatility of net asset returns. This part sets up multiple volatility models and makes a comprehensive evaluation to find out a relatively superior model, which is used to predict the volatility of net asset returns and to test the asymmetry in the volatility of net asset returns.
(5) an empirical study of the dynamic relationship between the volatility of net asset returns and the cash flow. This part first establishes a vector autoregressive model with the volatility of net asset returns and cash flow as a variable, and uses the Granger causality analysis and impulse response function to analyze the structure, and then changes in the net assets yield and cash flow. Vector auto regression model is established and variance analysis is used for structural analysis.
(6) conclusion. This part summarizes the research done in this paper, explains the existing problems, and points out the future research direction.
Through literature comparison and theoretical deduction, this paper draws a conclusion that the GARCH model is better for the fitting effect and prediction ability of the time series of financial data, and the better the prediction time limit is, the better the EGARCH model is in the unsymmetry effect in the capture of the volatility; it has realized the preview of the volatility rate to the high frequency transaction data. With the help of GARCH model and EGARCH (1,1) model, and several other basic GARCH expansion model --GARCH-M models, TGARCH model and PGARCH model, this paper empirically analyses the fluctuation of net asset returns. The results show that the fitting effect and prediction ability of GARCH (1,1) model to the net asset returns of Chinese listed companies are relative. The EGARCH (1,1) model confirms that there is a significant asymmetric effect in the volatility of net asset yield in China's listed companies.
An empirical study on the relationship between the volatility of net assets yield and cash flow is studied in this paper. In this paper, the VAR model is established with the volatility of net asset returns and the cash flow as variables, and the structural analysis is carried out by means of Granger causality analysis and impulse response function analysis. There is a two-way Granger causality between the gold flow, that is, the two parties are each other's Granger reasons, and the volatility of the net asset returns is more affected by the impact of their own residual impact than the residual of the cash flow, and the response is more rapid when they are affected by themselves. At the same time, the impact of the cash flow on its own residual impact is greater than the volatility of the net assets. Secondly, the VAR model is set up with the net asset returns and cash flow as variables, and the structural analysis is carried out by variance decomposition. The conclusion shows that the ratio of volatility of net assets yield fluctuation is far greater than that caused by the residual of cash flow, and the change of the variance of cash flow is contributed by its own residuals. The ratio is also greater, and this contribution ratio will change appropriately when the impact arises, and will remain stable in the future.
【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F224;F832.51

【参考文献】