Simulation Approach for Estimating the Parameters of the First Type Extreme Value Distribution

Abstract

In this paper, the two parameters of type I extreme value distribution are estimated for the maximum values by using moment’s method and the weighted least square method. The simulation approach is used to compare the obtained results of the applied methods in order to get the best method to estimate the parameters, in which the simulation process starts by generating random samples follows the extreme value distribution. This algorithm is based on three samples of the real parameters and with different sample sizes. The results are tabulated for comparison purpose in connection with the mean square error.