The Anderson-Darling Test Statistic of the Generalized Extreme Value

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 different 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.