基于VaR方法对我国可转债市场风险的实证研究

发布时间：2018-02-08 23:44

本文关键词： 可转换债券 GARCH族模型 VaR　出处：《首都经济贸易大学》2017年硕士论文　论文类型：学位论文

【摘要】：可转换公司债券以其兼具债性和股性的特殊结构优势越来越受到投资者欢迎,“向下债券保底,向上收益可期”的美誉也被市场中绝大多数的投资者所认可。然而,这种特殊的结构也使得可转债面临诸多复杂的风险因素,这些风险相互影响进而使得可转债的风险度量工作难度加大。基于此,本文以我国可转债市场为研究对象,利用市场的日交易数据和数学模型探索我国可转债市场的风险度量模型,在此基础上试图探寻我国可转债市场的总体风险水平,并对可转债和可交换债的风险特点进行研究。本论文共分为六章,前两章在介绍本文研究背景和意义的基础上利用收集到的数据对我国可转债的一级和二级市场特点进行总结,分析了可转债投资所面临的复杂投资风险并对重点风险进行重点介绍,明确了可转债风险度量工作的重要意义。第三章引入VaR模型,对VaR的不同计算方法进行综述并最终决定利用参数法来测度我国可转债的风险水平。同时引入基于不同分布的GARCH族模型,以更加精确的模拟可转债市场波动路径。第四章以我国中证转债指数2004年1月2日至2016年12月30日之间共计3158交易日的收盘数据为样本,在数据统计检验的基础上利用基于GARCH族模型的参数法测度VaR,并进行回测检验。最终发现基于t分布下GARCH族模型均会对市场风险高估,基于正态分布和GED分布下三种模型计算结果相近,除GED分布下EGARCH(2,2)模型的预测结果略高于5%之外,其他模型均能够较好的预测中证转债的市场风险。其中,从风险控制与管理角度,GED分布下的TGARCH模型预测效果最优,能够达到最优的的风险预测效果。第五章利用两组相同评级的可转债和可交换债数据尝试分析二者风险水平差异,实证结果证明可转债的总体风险水平要小于可交换债,因此投资可交换债更要对风险进行严格管控。最后在全文的研究基础上,本文进行系统总结并提出文章研究的不足之处。
[Abstract]:Convertible corporate bonds are becoming more and more popular among investors because of their special structural advantages of both debt and stock. The reputation of "keeping the bottom down and earning up" is also recognized by the vast majority of investors in the market. This special structure also makes convertible bonds face a lot of complex risk factors, which influence each other and make it more difficult to measure the risks of convertible bonds. Based on this, this paper takes China's convertible bond market as the research object. Based on the daily transaction data and mathematical model of the market, the paper explores the risk measurement model of China's convertible bond market, and then attempts to explore the overall risk level of China's convertible bond market. This paper is divided into six chapters. The first two chapters summarize the characteristics of the primary and secondary markets of China's convertible bonds on the basis of the background and significance of the research. This paper analyzes the complex investment risks faced by convertible bond investment, introduces the key risks, and clarifies the significance of the risk measurement of convertible bonds. Chapter three introduces the VaR model. This paper summarizes the different calculation methods of VaR and finally decides to use the parameter method to measure the risk level of convertible bonds in China. At the same time, the GARCH family model based on different distributions is introduced. Using a more accurate simulation of the volatility path in the convertible bond market. Chapter 4th is based on the closing data of China's China Securities Exchange Index from January 2nd 2004 to December 30th 2016 for a total of 3,158 trading days. On the basis of statistical test of data, the parameter method based on GARCH family model is used to measure VaR, and the back test is carried out. Finally, it is found that the GARCH family model based on t distribution will overestimate the market risk. Based on normal distribution and GED distribution, the calculation results of the three models are similar. Except for the GED distribution, the prediction results of EGARCHX 2 + 2) model are higher than 5%, and all the other models can better predict the market risk of securities to bonds. From the point of view of risk control and management, the TGARCH model under GED distribution has the best prediction effect and can achieve the optimal risk forecasting effect. Chapter 5th tries to analyze the difference of risk level between the two groups of convertible and exchangeable debt data with the same rating. The empirical results show that the overall risk level of convertible bonds is lower than that of exchangeable bonds, so the investment convertible bonds should be strictly controlled. Finally, on the basis of the research of the full text, this paper makes a systematic summary and puts forward the deficiencies of this paper.
【学位授予单位】：首都经济贸易大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F832.51

【参考文献】