Estimating Portfolio Risk with Conditional Joe-Clayton Copula: An Empirical Analysis with Asian Equity Markets

Financial modeling in developing markets requires dynamic and complex algorithms, which enable investors to estimate extremes in the returns arising from the chaotic characteristics of those markets. In this research paper, the value-at-risk (VaR) of a portfolio consists of Bombay Stock Exchange Index and Hang Seng Stock Exchange Index from January 02, 2002 to April 19, 2007 estimated by conditional Joe-Clayton copula. The performance of the model is compared with those of EWMA and Delta-Normal by using three different back-testing algorithms: Kupiec test (1995), Christoffersen test (1998) and Berkowitz test (2001). The empirical results show that conditional copula captures the extremes in return distribution of the portfolio more successfully by providing flexible joint distributions and splitting the marginality from the dependencies between the assets in the portfolio.