Estimating Dependence Structure and Risk of Financial Market Crash

Abstract

In this paper, we estimate the dependence structure between international stock markets using copulas. Different relationships that exist in normal and extreme periods were estimated using Clayton copula. The Inference Functions for Margins method was used in estimating the clayton copula parameter thereby obtaining dependence estimates used in estimating Value-at-Risk. Extreme events are likely to alter the dependence structure of financial markets.This could have implications for investment decisions and ability to estimate the risk of financial markets crash. Results reveal that during the crisis period (2007-2009), maximum possible loss of market value is 75.9% and 77.6% with a confidence interval of 90% for the Kenya-Nigeria and Kenya-South Africa portfolios respectively. This implies that the Kenya-South Africa portfolio has the highest risk.