Modeling and Volatility Analysis of Share Prices Using ARCH and GARCH Models

We identify and estimate the mean and variance components of the daily closing share prices using ARIMA-GARCH type models by explaining the volatility structure of the residuals obtained under the best suited mean models for the said series. The parameters of ARIMA type simple specifications are routinely anticipated by applying the OLS methodology but it has two disadvantages when the volatility or ARCH effect is present. The first problem may be the autocorrelation in error terms. To handle this unwanted situation the lagged dependent variables can be incorporated as independent variables in the mean equation. The other problem may be the presence of ARCH effect. This problem can be resolved by employing the ARCH or GARCH specifications so we have taken advantage of such type of models in our study.