Goodness-of-fit tests for the Weibull distribution based on the Laplace transform

Krit, Meryam

Journal de la société française de statistique, Tome 155 (2014) no. 3, pp. 135-151.

L’objectif de cet article est de développer de nouveaux tests d’adéquation à la loi de Weibull à deux paramètres basés sur la transformée de Laplace. Le principe de ces tests consiste à mesurer la proximité entre la transformée de Laplace théorique et sa version empirique. Trois méthodes d’estimation des paramètres de la loi de Weibull sont utilisées pour simplifier la construction des statistiques. L’article propose aussi une nouvelle version de la statistique de Cabaña et Quiroz utilisant les estimateurs de maximum de vraisemblance des paramètres. Ces tests ne sont pas asymptotiques, ils peuvent être utilisés pour des échantillons de petite taille. Une comparaison exhaustive des tests proposés est présentée. Parmi tous les tests d’adéquation utilisés, les meilleurs tests sont identifiés. Les résultats dépendent fortement de la forme du taux de hasard de la loi sous-jacente.

The aim of this paper is to develop new goodness-of-fit (GOF) tests for the two-parameter Weibull distribution based on the Laplace transform. The principle of the tests relies on the measure of the closeness between the theoretical Laplace transform and its empirical version. Three estimation methods are used to simplify the building of the statistics. The paper also introduces a new version of Cabaña and Quiroz statistic using the maximum likelihood estimators of the parameters. All these tests are not asymptotic and can be used for small samples size. A comprehensive comparison study is presented. Among all the proposed GOF tests, the best ones are identified. The results strongly depend on the shape of the underlying hazard rate.

[1] Antle, C.E.; Bain, L.J. A property of maximum likelihood estimators of location and scale parameters, SIAM Review, Volume 11 (1969) no. 2, pp. 251-253 | Zbl

[2] Ascher, S. A survey of tests for exponentiality, Communications in Statistics, Theory and Methods, Volume 19 (1990) no. 5, pp. 1811-1825

[3] Bousquet, N.; Bertholon, H.; Celeux, G. An alternative competing risk model to the Weibull distribution for modeling aging in lifetime data analysis, Lifetime Data Analysis, Volume 12 (2006), pp. 481-504 | Zbl

[4] Baringhaus, L.; Henze, N. A class of consistent tests for exponentiality based on the empirical Laplace transform, Annals of the Institute of Statistical Mathematics, Volume 43 (1991), pp. 551-564 | Zbl

[5] Cabaña, A.; Quiroz, A.J. Using the empirical moment generating function in testing the Weibull and type 1 extreme value distributions, Test, Volume 14 (2005) no. 2, pp. 417-431 | Zbl

[6] D’Agostino, R.B.; Stephens, M.A. Goodness-of-fit techniques, Marcel Dekker, 1986 | Zbl

[7] Henze, N. A new flexible class of omnibus tests for exponentiality, Communications in Statistics, Theory and Methods, Volume 22 (1993), pp. 115-133 | Zbl

[8] Henze, N.; Meintanis, S.G. Recent and classical tests for exponentiality : a partial review with comparisons, Metrika, Volume 61 (2005), pp. 29-45 | Zbl

[9] Henze, N.; Meintanis, S.G.; Ebner, B. Goodness-of-fit tests for the Gamma distribution based on the empirical Laplace transform, Communications in Statistics, Theory and Methods, Volume 41 (2012) no. 9, pp. 1543-1556 | Zbl

[10] Jimenez-Gamero, M.D.; Alba-Fernandez, V.; Munoz-Garcia, J.; Chalco-Cano, Y. Goodness-of-fit tests based on empirical characteristic functions, Computational Statistics and Data Analysis, Volume 53 (2009), pp. 3957-3971 | Zbl

[11] Krit, M.; Gaudoin, O. Review and comparison of goodness-of-fit tests for the exponential and Weibull distributions, 11th International Probabilistic Safety Assessment and Management Conference and the Annual European Safety and Reliability Conference 2012, PSAM11-ESREL 2012. Helsinki, Finlande (2012), pp. 1404-1413

[12] Krit, M.; Gaudoin, O.; Xie, M.; Remy, E. Simplified likelihood based goodness-of-fit tests for the Weibull distribution (submitted) | Zbl

[13] Liao, M.; Shimokawa, T. A new goodness-of-fit test for type-1 extreme-value and 2-parameter Weibull distributions with estimated parameters, Journal of Statistical Computation and Simulation, Volume 64 (1999) no. 1, pp. 23-48 | Zbl

[14] Mudholkar, G.S.; Srivastava, D.K. Exponentiated Weibull family for analyzing bathtub failure-rate data, IEEE Transactions on Reliability, Volume 42 (2) (1993), pp. 299-302 | Zbl

[15] Meintanis, S.G.; Swanepoel, J.; Allison, J. The probability weighted characteristic function and goodness-of-fit testing, Journal of Statistical Planning and Inference (2013) (http://dx.doi.org/10.1016/j.jspi.2013.09.011) | Zbl

[16] Mann, N.R.; Scheuer, E.M.; Fertig, K.W. A new goodness-of-fit test for the two-parameter Weibull or Extreme-value distribution, Communications in Statistics, Volume 2 (1973), pp. 383-400 | Zbl

[17] Murthy, D.N.P.; Xie, M.; Jiang, R. Weibull models, Wiley, 2004 | Zbl

[18] Rinne, H. The Weibull distribution -A handbook, CRC-Chapman and Hall, 2009 | Zbl

[19] Stacy, E.W. A generalization of the gamma distribution, Annals of Mathematical Statistics, Volume 33 (1962), pp. 1187-1192 | Zbl

[20] Tiku, M.L.; Singh, M. Testing the two-parameter Weibull distribution, Communications in Statistics, Volume 10 (1981), pp. 907-918

[21] Vaart, A. W. Van Der Asymptotic Statistics (in Statistical, Cambridge Series; Mathematics, Probabilistic, eds.), Cambridge university Press, 1998 | Zbl

[22] Xie, M.; Lai, C.D. Reliability analysis using additive Weibull model with bathtub-shaped failure rate function, Reliability Engineering and System Safety, Volume 52 (1995), pp. 87-93