基于DCC-GARCH模型的沪市时变贝塔及财务影响因素分析

发布时间：2018-06-28 11:22

本文选题：资本资产定价模型 + DCC-GARCH　；参考：《天津大学》2014年硕士论文

【摘要】：在中国股票市场的系统风险所占比例仍居高不下的情况下，，研究用于代表系统风险的贝塔系数的特征势在必行。到目前为止，国内学者在这一领域已经进行了一定的研究，取得了一定的理论成果。但通过对国内外研究的分析发现，虽然已有一部分国内学者将计量经济学模型应用于贝塔系数的估计，但一般采用Bollerslev提出的常相关系数模型来估计相关系数，该模型并不符合金融市场的实际情况。另一方面，国内对于会计变量与系统风险的关系的研究仍沿用贝塔系数为常数的假定，这与已有的研究结论认为贝塔系数具有不稳定性相矛盾，由此得到的研究结论将不可避免的存在一定的争议。本文基于计量经济学的相关理论，研究了上市公司系统风险与财务指标之间的相关关系，在实证分析中采用与现实市场情况更贴近的动态条件相关多元GARCH模型估计贝塔系数。同时，用时变贝塔代替常数贝塔来代表股票的系统风险并研究其与代表公司基本特征的财务指标之间的关系。目前，我国对于DCC-GARCH模型的实现步骤尚未形成完整的体系，因此本文重点解释了利用该模型估计时变贝塔的具体步骤，并形成了较为规范的体系。此外，按照时间顺序对系统风险与会计变量关系的理论模型进行了梳理，在总结已有研究的基础上引入新的变量作为系统风险影响因素的回归模型。本文证明了用来衡量资产风险的贝塔系数具有不稳定性，并且个股对市场的变化较为敏感。通过实证研究得到系统风险与财务指标之间的相关关系，即对由时变贝塔表示的系统风险有显著正向影响的财务指标包括经营杠杆、资本积累率、企业规模以及应收账款周转率；有显著负向影响的指标包括流动比率和现金流量比率；无显著影响的指标为财务杠杆和净资产收益率。本文从贝塔系数的不稳定性的角度入手，研究财务指标对于以时变贝塔为代表的上市公司系统风险的影响，为研究系统风险的影响因素提供了新的思路，填补了该领域的研究空白。同时为拓展贝塔系数的理论价值、完善上市公司信息披露制度、改进贝塔系数的估计方法和构建上市公司基于系统风险的影响因素的风险报告体系提供了理论指导。
[Abstract]:Under the situation that the proportion of systematic risk in Chinese stock market is still high, it is imperative to study the characteristics of Beta coefficient which is used to represent the system risk. Up to now, domestic scholars have carried out certain research in this field and obtained certain theoretical results. However, through the analysis of domestic and foreign studies, it is found that although some domestic scholars have applied econometrics model to the estimation of Beta coefficient, the usual correlation coefficient model proposed by Bollerslev is generally used to estimate the correlation coefficient. The model does not accord with the actual situation of the financial market. On the other hand, the domestic research on the relationship between accounting variables and system risk still follows the assumption that Beta coefficient is constant, which contradicts the previous research conclusion that Beta coefficient is unstable. The conclusion will inevitably be controversial. Based on the theory of econometrics, this paper studies the correlation between system risk and financial index of listed companies. In the empirical analysis, the dynamic condition correlation GARCH model, which is closer to the real market situation, is used to estimate Beta coefficient. At the same time, the time-varying beta is used instead of the constant beta to represent the systematic risk of the stock and to study the relationship between the time-varying beta and the financial indexes which represent the basic characteristics of the company. At present, the implementation steps of the DCC-GARCH model in China have not yet formed a complete system, so this paper mainly explains the concrete steps to estimate the time-varying beta using the model, and forms a more standard system. In addition, the theoretical model of the relationship between system risk and accounting variables is sorted out according to the time order, and a new variable is introduced as the regression model of the influencing factors of system risk on the basis of summarizing the existing research. In this paper, it is proved that the beta coefficient used to measure asset risk is unstable and that individual stocks are sensitive to market changes. Through the empirical study, the correlation between system risk and financial indicators is obtained, that is, the financial indicators which have significant positive impact on the system risk expressed by time-varying beta include operating lever, capital accumulation rate, and so on. Enterprise size and turnover rate of accounts receivable; the indicators with significant negative impact include current ratio and cash flow ratio; the indicators with no significant impact are financial leverage and return on net assets. From the point of view of the instability of beta coefficient, this paper studies the influence of financial indexes on the system risk of listed companies represented by time-varying beta, which provides a new way to study the influencing factors of system risk. It fills up the research gap in this field. At the same time, it provides theoretical guidance for expanding the theoretical value of beta coefficient, perfecting the information disclosure system of listed companies, improving the estimation method of beta coefficient and constructing the risk reporting system of listed companies based on the influencing factors of system risk.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51

【参考文献】