Volatility in frontier markets: a Multivariate GARCH analysis

Abstract

Purpose The purpose of this paper is to analyze the existence of volatility spillover effect in frontier markets. This study also examines whether any linkages exist among these markets or not. Design/methodology/approach Monthly data of regional frontier markets, from 2009 to 2016, are analyzed using Multivariate GARCH (BEKK and Dynamic Conditional Correlation (DCC)) models. Findings The result of cointegration test shows that the sample frontier markets are not linked in long run, and Granger causality test reveals that the markets under consideration do not cause each other even in the short run. BEKK test says that the effect of the arrival of shock from the own market does not last for longer, whereas shock from other markets lasts with the stronger persistence, and according to DCC test, the volatility spillover exists for all the markets. Practical implications The results of present study suggest that the frontier markets are not cointegrated in the long run as well as in the short run, which opens the doors for long-term investments in these markets in future, which may lead to decent returns. Long-term investors may draw the benefits from including the financial assets in their portfolios from these non-integrated frontier markets; nevertheless, they have to consider and implement diversification and hedging strategies during the period of financial turmoil, so as to protect themselves against economic and financial distress. Originality/value Significant work has been done on developed, developing and emerging markets but frontier markets are not explored much so far. This paper is an attempt to see the status of frontier stock markets as potential financial markets for diversification benefits.