A Copula-Based Quantile-on-Quantile Regression Approach to Modeling Dependence Structure between Stock and Bond Returns: Evidence from Historical Data of India, South Africa, UK and US

The current market environment presents a host of challenges to investors that require using a new approach to portfolio construction, particularly when it comes to diversification. The correlation structure across assets is a key feature of the portfolio choice problem because it determines the riskiness of the investment position. The present paper uses copula quantile-on-quantile regression (C-QQR) to model correlation structure. Even though this technique is driven by copula-based quantile regression model, it retains more flexibility and thus, should deliver more robust and accurate estimates. In particular, we investigate the entire dependence structure of stocks and bonds in the cases of India, South Africa, UK and US by using their return information to characterize the environments of the respective financial markets. The former are two important emerging markets, while the latter two of the most mature ones globally. Dependence in bear, bull and normal states of stock (bond) markets can be summarized by the lower, upper and centrally located quantiles of the respective stock (bond) returns. Based on historical monthly data covering the periods of 1920:08-2017:02, 1910:01-2017:02, 1933:01-2017:02, and 1791:09-2017:02 for India, South Africa, UK and the US respectively, it was shown that there is substantial heterogeneity in the bond-stock returns correlation across the countries under study. Additionally, the findings reported herein suggest that using C-QQR in portfolio management can enable the formation of tailored response strategies, adapted to the needs of investors and traders.