Volatility Modeling and Forecasting: A Study of Frontier South Asian Markets



Modeling and forecasting volatility have gained much interest among researchers as the use of volatility became widespread in financial data analysis. Several studies have been carried out to model stock market volatility in various countries. This paper employs the GARCH-family models to study the volatility of stock market returns of the frontier South Asian markets that include Sri Lanka, Pakistan, Vietnam, and Bangladesh. The objective of the study is to determine the best fitting GARCH-family models based on Akaike?s information criterion (AIC) for the frontier markets and assess their forecasting performance. The paper estimated the GARCH (1,1), GARCH-M (1,1), EGARCH (1,1) and TGARCH (1,1,1) models to capture the volatility of the frontier markets and subsequently their forecasting performances were evaluated using statistical loss functions. Appropriate models have been utilized to forecast volatility and forecasting accuracy were evaluated using root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE). The empirical findings show that modeling the asymmetric properties of volatility is important as the asymmetric models outperformed their symmetric counterparts. More specifically, the TGARCH (1,1,1) was found to be the best fitted model and the best performing model for both Sri Lanka and Pakistan. Nevertheless, the EGARCH (1,1) was found to be the best fitted model and the TGARCH (1,1,1) was the best performing model for both Vietnam and Bangladesh. Keywords: Volatility, GARCH, EGARCH, TGARCH, South Asian Markets

 Artículos similares