基于Copula-Jump-GARCH模型的借贷资产组合优化

发布时间：2018-01-03 16:09

本文关键词：基于Copula-Jump-GARCH模型的借贷资产组合优化　出处：《南京财经大学》2013年硕士论文　论文类型：学位论文

更多相关文章： Jump-GARCH模型 Copula函数 CVaR

【摘要】：按照均值-方差思想，资产组合优化就是找到收益率一定风险最小的资产组合或者是一定风险下的收益率最大的资产组合。传统的均值-方差理论同时假设单个资产和联合分布都服从正态分布。但是大量的实证研究结果表明资产的收益率往往表现出波动性聚类、尖峰厚尾性和跳跃性等特点，而资产之间的相关关系也呈现出非线性。因此用传统的均值-方差模型来寻找资产组合的最优化策略已经不符合现实。为了克服传统理论的不足，本文主要运用Jump-GARCH模型来拟合单个资产的收益率，用Copula函数来刻画资产之间的相关结构关系，并把两者相结合共同描述资产组合的联合分布，用Mean-CVaR方法来解决借贷资产组合优化问题，通过计算一定期望收益率下的最小CVaR值得到最优借贷资产配置策略。本文分别用GARCH、SV和Jump-GARCH模型对上证综合指数进行拟合对比分析，发现Jump-GARCH模型能够更好地拟合资产收益率的各种特性。用Jump-GARCH模型对中国联通、中国石化和中青旅这三家公司的收益率进行研究，并通过对比刻画三家公司之间相关结构的几种Copula函数，最终得到了t-Copula函数能够更好地构建资产组合之间的相关结构。最后在Mean-CVaR模型的限制下，得到了借贷资产的最优配置策略。这给银行在决策投放贷款比例时，，提供了一定的理论指导。
[Abstract]:According to the mean-variance idea. Portfolio optimization is to find the portfolio with the minimum return rate or the largest yield under a certain risk. The traditional mean-variance theory assumes that both the individual asset and the joint distribution are conformable to each other. But a large number of empirical results show that the return rate of assets tends to show volatility clustering. The correlation between assets is nonlinear, so the traditional mean-variance model is not in line with the reality. In order to overcome the shortcomings of traditional theory, this paper mainly uses Jump-GARCH model to fit the return rate of a single asset, and uses Copula function to describe the relationship between assets. And the combination of the two to describe the joint distribution of portfolio, using the Mean-CVaR method to solve the loan portfolio optimization problem. By calculating the minimum CVaR value under a certain expected rate of return, the optimal loan asset allocation strategy is obtained. In this paper, GARCHN SV and Jump-GARCH models are used to compare and analyze Shanghai Composite Index. It is found that the Jump-GARCH model can better fit the various characteristics of the return on assets. The Jump-GARCH model is applied to China Unicom. The return rate of Sinopec and China Youth Tours is studied, and several Copula functions are compared to describe the correlation structure between the three companies. Finally, the t-Copula function can better construct the correlation structure between asset combinations. Finally, under the limitation of Mean-CVaR model. The optimal allocation strategy of loan assets is obtained, which provides certain theoretical guidance for banks to make decision on loan ratio.
【学位授予单位】：南京财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F224;F832.51

【参考文献】