Time series modeling of KSE-100 index

Abstract

The study aims to establish the Generalized Auto Regressive Conditional Heteroskedasticity (GARCH) model, to analyze stock returns of KS-100 Index of Pakistan Stock Exchange. After applying different GARCH models, this study attempted to select the best model. The daily values of the KSE 100 Index of Pakistan Stock Exchange were collected over the period of 4 years and 9 months, starting from December 2013 to September 2017. The study employed 1200 total observations. The objective was achieved by estimating different Auto Regressive Conditional Heteroskedasticity (ARCH) model and its extension, GARCH model, identification of ARIMA model, estimation of identified ARIMA and then checking it. The goodness of ut was assessed through the Akaike Information Criteria (AIC), Schwarz Information Criteria (SIC)and Standard Error(S.E). Moreover, the parameters of the model were estimated, using maximum likelihood method. The present study concludes that GARCH (1, 1) is the most suitable model, to capture the volatility of stock returns of KSE-100 Index and vital for academicians and econometricians. It is also helpful for the policymakers to know the trend and patterns of stock exchange.