Volatility Estimation Using Symmetric and Asymmetric Models in Oil Exporting Emerging Markets

Abstract

The study empirically investigates the volatility pattern of thirteen emerging economies which are predominantly oil exporting countries. It is based on the time series data which consists of monthly closing price data of their index for a ten-year period from 01 January 2008 to 31 December 2017. Emerging markets are considered as investment destinations due to the presence of risk premium which has made the stock markets of these countries more volatile. Added to this is that these countries underwent crisis due to the sharp decline in crude oil prices as they were primarily dependent on oil exports. Hence it is a significant to study the volatility behavior of these countries. The study has been done by employing both symmetric and asymmetric models of generalized autoregressive conditional heteroscedastic. As per Akaike Information Criterion (AIC), Log likelihood and Schwarz Information Criterion (SIC) the study provides evidence that GARCH (1,1) and TGARCH(1,1) estimations are found to be the most appropriate model that fits symmetric and asymmetric volatility respectively for all the thirteen countries. There was evidence of volatility clustering and leptokurtic in all the countries considered in the study. While EGARCH model revealed no support of existence of leverage on the stock returns, TGARCH supported existence of leverage in case of four countries. The tests for asymmetries in volatility indicate the size effect of the news, reaffirmed through the results of sign bias tests and news impact curves, which indicate that the size effect is stronger for bad news than the good news for countries which supported existence of leverage.