The Measurement of Corporate Default Correlation Based on Optimal Copula Function

摘要

Copula function is a function that links joint distribution function of random vectors and its corresponding components marginal distribution function. It is an important tool to describe the structure relationship between many financial markets. With using the Copula function to construct the joint distribution function, it will not be limited by the marginal distribution function, instead, the marginal distribution function of random vectors and its dependency structure can be studied separately. The key to construct corporate default correlation by Copula function is that select a suitable Copula function from numerous Copula functions. It chooses the optimum Copula function by graphic method and minimum variance test method. The original analysis data is Haier's and Midea electronics appliance's logarithmic return rate of stock day from April 31th, 2005 to April 31th, 2009. In the end, it describes default correlation between the two enterprises by tail correlation coefficient.