DOI: 10.36347/sjpms.2024.v11i02.001

Pages: 18-26

Extreme value theory (EVT) is one of major importance in many fields of applications where extreme values may appear and have detrimental effects and finally Generalized Extreme Value (GEV) distribution model is found as the best fitted distribution model. Extreme minimum temperature using 40 years of data is studied. Minimum of four different time periods (monthly, quarterly, half yearly and yearly) are fitted to the minimum Generalized Extreme Value (MGEV) distribution. The first objective of this study is to describe and model the behaviour of extreme minimum temperature in Sebha by using the MGEV distribution. The second object, we determine the effect of using different size blocks, and illustrate the best four different time (Monthly, Quarterly, Half yearly and yearly) selection periods that are suitable for modelling with the MGEV distribution. The method of probability-weighted moments (PWM) has been used to estimate the unknown parameters; and its corresponding Deviance Test (DT) approach to test the goodness of fit. The results show that minimum Weibull distribution as special case of MGEV distribution is the most an appropriate choice of all selection periods (quarterly half yearly and yearly) are significant except only monthly of all three periods are significant to be fitted by minimum Frechet model. We will use the R programming with packages of fExtreme, evir and ismev to calculate, parameter estimation, testing and diagnostic plots.