A Study on Dependence Structure Between Bond Yields and Stock Prices During Lockdowns Using Bivariate Copula

Abstract

A copula is a multivariate cumulative distribution function in probability theory and statistics for which the marginal probability distribution of each variable is uniform on the range [0, 1]. Copulas are used to describe and model the inter-correlation or dependency between random variables. This research aims to study a copula-GARCH approach to modelling joint distribution of two major assets which is stock price and ten-year Treasury bonds yields during lockdowns. Our objectives is to analyse the dependence structure of two major asset classes between bonds yields and stock price and to investigate the effect of lockdowns to the bond yields and stock prices. R-studio is used for modelling the distribution, and fit the data with Student-t distribution copula and Clayton copula and estimate the tail dependence. Our result show that Student-t distribution copula yield higher log-likelihoods while Clayton copula yield lower log-likelihoods. Other than that, stock prices and bond yields is effected during lockdowns. Our result show that’s the correlation between stock price and bond yields is increase overtime during movement control order.