我国房地产行业公司债券信用利差定价模型研究

发布时间：2018-04-01 20:28

本文选题：房地产　切入点：公司债券　出处：《华东师范大学》2013年硕士论文

【摘要】：经过6年的发展,公司债券渐渐成为越来越多经营者降低融资成本的外部融资工具,也成为了越来越多投资者追捧的对象。因而公司债券的风险补偿,即信用利差成为了发行人和投资者进行投融资时最为关注的问题。本文选取了房地产市场公司债券作为研究对象,基于对时间序列分析的方法,通过分析我国公司债券的宏微观影响因子,提供了一种建立了二级市场上房地产行业公司债券信用利差预测模型的方法。首先,从债券自身因素、宏观经济因素和企业财务因素三个层面来考虑信用利差的影响因素。其次,利用聚类分析的方法从纳入考虑的众多影响因子中选取了7个宏微观因子,建立信用利差预测模型,这是本文的重点。这一步中,采用多项式拟合的方法调整了财务数据频率,利用逐步回归的方法排除非显著影响因子,建立信用利差与显著影响因子之间的协整关系后建立残差的ARMA模型对协整关系进行修正,得到了第一类信用利差预测模型。接着又将非显著因子引入到模型中,统一了行业中所有债券信用利差影响因子,建立了第二类信用利差预测模型。经过对比,两类模型在预测效果上的差别非常细微,但第二类模型在实务操作中有更高的便利性和可比性。最后,本文将第二类模型与已有的信用利差模型进行对比,论证了本文第二类模型的合理性和优势。通过上述研究过程,本文得到了如下研究结论：对房地产行业公司债券信用利差产生影响的微观因子有剩余到期期限、总资产、净利润和经营活动现金净流量,宏观因素有股票市场收益率、利率期限结构斜率和工业增加值。对不同债券信用利差产生显著影响的因子是不一样的,但引入非显著因子并不会影响模型的预测效果。相反,引入非显著因子可以提高实务操作中的便利性和可比性。
[Abstract]:After six years of development, corporate bonds have gradually become an external financing tool for more and more operators to reduce their financing costs, and become more and more popular among investors. That is, credit spreads have become the most concerned issue when issuers and investors invest and finance. This paper selects corporate bonds in real estate market as the research object, based on the method of time series analysis, through the analysis of the macro and micro factors of corporate bonds in China. This paper provides a method for predicting the credit spreads of real estate companies in the secondary market. The influence factors of credit spreads are considered from three aspects of macroeconomic factors and enterprise financial factors. Secondly, seven macro and micro factors are selected from the many factors taken into account by cluster analysis, and a credit spread prediction model is established. This is the focus of this paper. In this step, the frequency of financial data is adjusted by polynomial fitting method, and the non-significant influence factors are excluded by the method of stepwise regression. After establishing the cointegration relationship between credit interest difference and significant influence factor, the ARMA model of residual error is established to modify the co-integration relation, and the first kind of credit interest rate prediction model is obtained. Then, the non-significant factor is introduced into the model. Unifies the influence factors of all bond credit spreads in the industry, and establishes the second kind of credit spread forecasting model. After comparison, the difference between the two kinds of models in the forecasting effect is very small. But the second type model is more convenient and comparable in practical operation. Finally, this paper compares the second type model with the existing credit spread model, and proves the rationality and advantage of the second type model. Through the above research process, this paper obtains the following conclusions: the micro factors that affect the credit spreads of real estate companies are the remaining maturity period, total assets, net profit and net cash flow of operating activities. Macro factors include stock market yield, interest rate term structure slope and industrial added value. The factors that have significant influence on different bond credit spreads are different, but the introduction of non-significant factors will not affect the forecasting effect of the model. The introduction of non-significant factors can improve convenience and comparability in practical operations.
【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F224;F299.23;F832.51

【参考文献】