Combination Of Investment CAPM-SV-Copula Model

Since Markowitz’s mean-variance portfolio selection theory was proposed, it was highly regarded in the field of economics and financial researches and it had made a series of deep theoretical results and extensive practical applications.As we all know, Markowitz’s mean-variance model is built dependently on some assumptions, such as that the distributions of the expected return of assets are known, independently and identically distributed and that it meets the maximization of single utility to the investor and so on. But with the change and development of economic environments, these assumptions are idealistic or difficult to meet. Therefore, it is natural to find a generalized portfolio model and an analysis method (it may differ from Markowitz Linear income portfolio) which is more reasonable, more effective and more suitable to the financial market. It can rich the portfolio theory as well as that it has practical applications.The GARCH-Copula model even the SV-Copula model offers ideas and tools for the realization of the generalized portfolio. They provide an initially feasible solution to portfolio management which can solve the problem that the distributions of the assets are non-independent and non-identically distributed. However, it may be that the mathematical analysis of these two models are difficult, these two models have ignored a basic fact that the broad portfolio is influenced by market factors. Thus, there is room for further improvement. In view of this, this article will add the market factor to traditional SV model to construct a generalized CAPM-SV model. We use it to measure the financial risk and provide the investors with a more rational analysis tool.First, considering the impact of market, we combine it with Copula theory to construct a generalized CAPM-SV model. Then through the analysis of the volatility of the Industrial and Commercial Bank of China’s returns, we found that the model which considers the impact of market factors is better than the original one, because it can fit the data better and the risk prediction ability is better.Finally, we combine the generalized CAPM-SV model with Copula theory to construct the CAPM-SV-Copula model. Through the empirical analysis of the portfolio of Industrial and Commercial Bank of China and China National Petroleum Corporation, we found that the effect of the improved model in this paper is better than the traditional SV-Copula model in the ability to characterize the risk, particularly to describe the tail risk.