Pengujian statistik anderson-darling bagi taburan nilai ekstrim teritlak

Abstract

This paper presents the results of developing and evaluating goodness-of-fit tests for the generalized extreme value (GEV). The critical values tables AD test are developed based on the Pearsons chi-squared and the likelihood ratio statistic for GEV distribution is developed through Monte-Carlo simulation. The critical values for the statistics are modeled through regression to provide equations to estimate the critical values for any shape parameters distribution. Simulation studies using several di®erent distributional forms show that the critical values tables AD test based on the likelihood ratio statistic has a larger power than the AD test based on the Pearson chi-squared for testing the GEV distribution.