The economic determinants of comovement across African stock markets

Abstract

The aim of this study is to investigate the economic forces behind African stock markets interdependence using the recent and high frequency data, which spans a large period from 2004 to 2014 this by following two steps. First, we applied DCC-GARCH model to extract the dynamic conditional correlations between the selected markets, which are used as proxies measuring the stock markets interdependence. Second, we regress all the possible "time varying correlations" on potential economic variables, and then, estimate the regressions in two different ways, first as a pooled sample; second, as a system of seemingly unrelated regression (SUR) analysis. The results indicate the existence of trend toward more stock market interdependence as all possible pairs of stock markets interact significantly over time especially after the recent financial crisis. This result is confirmed by the pooled regression and SUR analysis, which suggest positive coefficient of the time trend and the 2008-crisis's dummy variable. The results of the pooled regression show that in addition to the contagion effect, a significant proportion (more than 36%) of the comovement among stock markets could be explained by economic factors, as several variables are significantly associated with the evolution of stock market comovement over time, some factors promote the interdependence some others depress it.