Abstract
The test for exponentiality of a dataset in terms of a specific aging property constitutes an interesting problem in reliability analysis. To this end, a wide variety of tests are proposed in the literature. In this paper, the excess-wealth function is recalled and new asymptotic properties are studied. By using the characterization of the exponential distribution based on the excess-wealth function, a new exponentiality test is proposed. Through simulation techniques, it is shown that this new test works well on small sample sizes. The exact null distribution and normality asymptotic is also obtained for the statistic proposed. This test and a new empirical graph based on the excess-wealth function are applied to extreme-value examples.
Similar content being viewed by others
References
Alonso AM, de-Zea-Bermudez P, Scotto MG (2014) Comparing generalized Pareto models fitted to extreme observations: an application to the largest temperatures in Spain. Stoch Environ Res Risk Assess 28:1221–1233
Anis MZ, Dutta S (2010) Recent tests of exponentiality against IFR alternatives: a survey. J Stat Comput Simul 80(12):1373–1387
Anis MZ, Mitra M (2011) A generalized Hollander–Proschan type test for NBUE alternatives. Stat Probab Lett 81:126–132
Bahadur RR (1966) A note on quantiles in large samples. Ann Math Stat 37:577–580
Balkema AA, de Haan L (1974) Residual lifetime at great age. Ann Probab 2:792–804
Borokov AA (1984) Estadística Matemática. Editorial Mir, Moscú
Castillo E, Hadi A (1997) Fitting the generalized Pareto distribution to data. J Am Stat Assoc 92(440):1609–1620
Chaouche M, Hubert P, Lang G (2002) Graphical characterisation of probability distribution tails. Stoch Environ Res Risk Assess 16:342–357
Dempster AP, Kleyle RM (1968) Distributions determined by cutting a simplex with hyperplane. Ann Math Stat 39:1473–1478
Embrechts P, Klüppelberg C, Mikosch T (1997) Modeling extremal events. Springer, New York
Fernández-Ponce JM, Infante-Macías R, Muñoz-Pérez J (1996) Characterization of lifetime distributions based on a quantile dispersion measure. Comput Stat Data Anal 21(5):547–561
Fernández-Ponce JM, Kochar SC, Muñoz-Perez J (1998) Partial orderings of distributions based on right spread function. J Appl Probab 35:221–228
Fernández-Ponce JM, Pellerey F, Rodríguez-Griñolo MR (2011a) On a new NBUE property in multivariate sense: an application. Comput Stat Data Anal 55(12):3283–3294
Fernández-Ponce JM, Pellerey F, Rodríguez-Griñolo MR (2011b) A characterization of the multivariate excess wealth ordering. Insurance 49(3):410–417
Gail MH, Gastwirth JL (1978) A scale-free goodness-of-fit test for the exponential distribution based on the Gini statistic. J R Stat Soc Ser B 40(3):350–357
Hoeffding W (1949) A class of statistics with asymptotically normal distribution. Ann Math Stat 19:293–295
Hollander M, Proschan F (1975) Tests for the mean residual life. Biometrika 62(3):585–600
Hosking JMR, Wallis JR (1987) Parameter and quantile estimation for the generalized Pareto distribution. Technometrics 29:339–349
Karr AF (1993) Probability. Springer, New York
Kochar SC, Wiens DP (1987) Partial orderings of life distributions with respect to their aging properties. Nav Res Logist Q 34:823–829
Kochar S, Li X, Shaked M (2002) The total time on test transform and the excess wealth stochastic orders of distributions. Adv Appl Probab 34(4):826–845
Kozubowski TJ, Panorska AK, Qeadan F, Gershunov A, Rominger D (2009) Testing exponentiality versus Pareto distribution via likelihood ratio. Commun Stat 38:118–139
Lekina A, Chebana F, Ouarda TBMJ (2014) Weighted estimate of extreme quantile: an application to the estimation of high flood return periods. Stoch Environ Res Risk Assess 28:147–165
Pickands J (1975) Statistical inference using extreme order statistics. Ann Stat 3:119–131
Pham TG, Turkkan N (1994) The Lorenz and the scaled total-time-on-test transform curves: a unified approach. IEEE Trans Reliab 43(1):76–84
Salvadori G (2003) Linear combinations of order statistics to estimate the quantiles of generalized Pareto distribution and extreme value distributions. Stoch Environ Res Risk Assess 17:116–140
Shaked M, Shanthikumar G (1998) Two variability orders. Probab Eng Inf Sci 12(1):1–23
Shaked M, Shanthikumar G (2007) Stochastic orders. Springer, New York
Silva AT, Portela MM, Naghettini M (2014) On peaks-over-threshold modeling of floods with zero-inflated Poisson arrivals under stationary and nonstationary. Stoch Environ Res Risk Assess 28:1587–1599
Smith RL (1987) Estimating tails of probability distributions. Ann Stat 15:1174–1207
Villaseñor-Alva JA, González-Estrada E (2009) A bootstrap goodness of fit test for the generalized Pareto distribution. Comput Stat Data Anal 53:3835–3841
Acknowledgments
The authors wish to thank the Editor and the two anonymous referees whose comments helped to improve a previous version of this paper. This work was partly supported by a grant of Junta de Andalucía (Spain) for research group (FQM-328).
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
In this section, we give the \({\mathbf{R}}\) code to compute critical values of our test and the estimated power for different sample sizes and parameter values for the Weibull distribution. In order to compute the powers for any other distribution, the corresponding R command should be changed.
Code 1. Computation of Critical Values for Statistic \(\Delta (F_{n})\) by using 10,000 MC samples.
Code 2. Estimated Power of the exponentiality test based on the excess-wealth function for the Weibull distribution by using 10,000 MC samples for a significance level equal to 0.05.
Rights and permissions
About this article
Cite this article
Fernández-Ponce, J.M., Rodríguez-Griñolo, M.R. Testing exponentiality against NBUE distributions with an application in environmental extremes. Stoch Environ Res Risk Assess 29, 679–692 (2015). https://doi.org/10.1007/s00477-014-0981-5
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-014-0981-5