ANALYZING PERFORMANCE OF GARCH MODELS IN NSE

The study uses three different models: GARCH(1,1), EGARCH(1,1) and GJR-GARCH(1,1) to analyze volatility of Nifty of National Stock Exchange (NSE) of India from January 1, 2010 to July 4, 2014. The results reveal persistence of volatility andthe presence of leverage effect implying impact of good and bad news is not same. To evaluate the models, various model selection and forecasting performance criterion like AIC, SBC, RMSE, MAE, MAPE and TIC criterionare employed. Our results indicate that GARCH (1,1) has better forecasting ability in NSE. JEL Classification: G14, C32 Key words: Volatility clustering, GARCH, EGARCH, TGARCH, RMSE, MAE, MAPE, TIC