Extreme Value Distributions on Closing Quotations and Returns of Islamabad Stock Exchange

Abstract

This study is an experimental test done on the secondary data of banking sector of Islamabad Stock Exchange for year 2017 and applied different techniques on the given data record by using Generalized Extreme Value Distribution (GEV), Gumble Distribution (GBL), Generalized Pareto Distribution (GPD), Exponential Distribution (EXP), Gamma Distribution (GAM), Weibull Distribution (WBL) on the data of four banks Habib Bank, Allied Bank, Bank Alfalah and Askari Bank. This data is concerning the closing quotations and returns of four banks registered in Islamabad Stock Exchange. We try to fit different distributions on the data and founnd the best fit distribution.
 We estimated the parameters of each distribution and also find the standard deviations of each distribution by using R Language and find which distribution is the best fit distribution on the basis of standard deviation distribution. We analyzed that shape wise GEV is the most suitable distribution, scale wise EXP distribution the best and location wise the best one is Gumbal distribution. This article shows that the overall GEV is the best distribution to model correctly the data.

Calibrated EWMA estimators for time-scaled surveys with diverse applications

“…It is worth mentioning that HBL is considered as stratum-I and UBL is considered as stratum-II. For more details about data, see references [33] , [34] . Some key characteristics are: Stratum-I , , , , , .…”