The Bond Market Correlation And Value At Risk Analysis Based On The Function Of Copula

At present, with China’s rapid economic development, the fiscal revenue and the fiscal policy’s macro-control role are also growing. The country’s infrastructure has made significant achievements. At the same time, The financial risks are gradually revealed. Pressures have been increasing on state expenditure, treasury bonds have been expanding, and the local government’s debts burden have also been heavy. Besides, with the inherent defects in the financial system and uncertain international political and economic factors, make the further expansion of financial risk. And how to prevent financial risks and maintain financial stability and sustainability, has become a subject of importance.Current financial risk research is major in characteristic analysis, quantitative analysis is very few, and most use the overall measure of financial risk indicators to measure the central government’s financial risk. The study lacks of micro-circulation of bonds accurate measure of market risk. Even there have some research in this area, most of them are using traditional linear methods to analyze the risks of associated variables. However, this has led to some problems, such as the normal distribution assumption is reasonable? Traditional linear correlation coefficient can reflect the dependencies of these variables? How to accurately measure the risk of a combination of variable values? These problems for China’s bond market risk management is of great significance, and thus to a more comprehensive understanding of China’s financial risk, also provide some useful measures and suggestions for the prevention of financial risk.In this paper, we use the Copula function in extreme value theory to measure the tail-related-risks of National bond market and corporate bond market, and compares with the obtained traditional linear method of correlation coefficients. Drawn:the linear correlation coefficient underestimated the systemic risk of the market, there are a lot of extreme value risk in the market.Then, the two market return series are fitted with generalized Pareto distribution, and Copula function-based Monte Carlo simulation is used to calculate the weights of different assets and confidence levels of the market value of portfolio risks, promptly the combination of VaR (Value at risk). Finally, for China’s current debt risk, gives some policies and recommendations.