Goodness-of-fit tests for the Weibull and Extreme Value distributions: a review and a comparative study

Olivier Gaudoin (1) , Meryam Krit (1) , Emmanuel Remy (2)

1 FIGAL - Fiabilité et Géométrie Aléatoire
2 EDF R&D MRI - Management des Risques Industriels

This paper presents an up-to-date review of Goodness-of-Fit (GoF) tests for the two-parameter Weibull and Extreme Value distributions. This review includes classical tests and more recent ones. An extensive comparison study is presented, which allows to identify the most powerful tests. The comparison is made using the EWGoF R package. Some guidance is given on the selection of the most appropriate tests based on the features of the data studied. Finally, an application to industrial data is presented.

Mots clés: Goodness-of-fit tests; Weibull distribution; Extreme Value distribution; Reliability

Domaines: Statistiques [math.ST]

Olivier Gaudoin, Meryam Krit, Emmanuel Remy. Goodness-of-fit tests for the Weibull and Extreme Value distributions: a review and a comparative study. 1st Euro-Mediterranean Conference on Mathematical Reliability, ECMR 2018, Jul 2018, Djerba, Tunisia. ⟨hal-01883084⟩