| dc.description.abstract | The purpose of the study is to model the volatility of the world stock indices. The word volatility is derived as the results of unequal variances of the error terms of the return series. The technical term used here is Heteroscedasticity. The investors in financial market consider the movement of the volatility of several locations for the risk management, derivatives pricing and hedging, market making, portfolio and for the different financial aims .The volatility modelling and forecasting is not very common in the Sri Lankan aspect. Volatility modelling with great emphasize on Sri Lanka is a very noteworthy attempt. The identification of the volatility models for the stock indices of Colombo stock exchange and stock markets of United States, India and United Kingdom are highly useful to the investors. The purpose of modelling the volatility of these selected stock indices is that Investors try to invest in several stock markets to diversify the risk. To capture the characteristics of volatility in the stock price series, Generalized Autoregressive Conditional Heteroscedasticity (GARCH) models have been used. Many GARCH family models were considered in this study. All Share Price Index of Colombo stock exchanges (ASPI), S & P 500 index of New York stock exchange, FTSE 100 of the London stock exchange and BSE SENSEX index of Bombe stock exchange have been considered in this study and the study period is from 1st January 2004 to 1st January 2014. GARCH (1, 1) model was identified as the best model for the ASPI return series, EGARCH (1,1) model was identified as the best model for both the FTSE 100 and BSESENSEX indices return series and while S & P 500 return series is best expressed by EGARCH (2,1) model. The model adequacy of the selected models have been tested using the ARCH LM test, Correlogram of squared returns and Correlogram of standardized residuals, while Q-Q plot was applied to check the error distribution. After the investigation of the numerical accuracy of the model estimated, the models have been used to forecast future volatility. | en_US |
