Bahman TARVİRDİZADE, Nader NEMATOLLAHİ

The power-linear hazard rate distribution andestimation of its parameters under progressivelytype-II censoring

In this paper, we introduce a class of distributions which generalizesthe power hazard rate distribution and is obtained by combining thelinear and power hazard rate functions. This class of distributions,which is called the power-linear hazard rate distribution, is simpleand flexible and contains some important lifetime distributions. Themaximum likelihood estimators of the parameters using the Newton-Raphson (NR) and the expectation-maximization (EM) algorithms andthe Bayes estimators of the parameters under squared error loss (SEL),linear-exponential (LINEX) and Stein loss functions are obtained basedon progressively type-II censored sample. Also, we obtain the asymp-totic confidence interval and some bootstrap confidence intervals andconstruct the HPD credible interval for the parameters. A real data setis analyzed and observed that the present hazard rate distribution canprovide a better fit than other three-parameter distributions. Finally, aMonte Carlo simulation study is conducted to investigate and comparethe performance of different types of estimators presented in this paper.

PDF

___

[1]Akdam, N., Kinaci, I. and Saracoglu, B.Statistical inference of stress-strength reliability forthe exponential power distribution based on progressive type-II censored samples, HacettepeJournal of Mathematics and Statistics46, 239–253, 2017.
[2]Bain, L.Analysis for the linear failure-rate life testing distribution, Technometrics15, 551–559, 1974.
[3]Balakrishnan, N. and Aggarwala, R.Progressive Censoring, Theory, Methods and Applica-tions, Birkhauser, Boston, 2000.
[4]Balakrishnan, N. and Sandhu, R.A.A simple simulation algorithm for generating progres-sively type-II generated samples, American Statistician49, 229–230, 1995.
[5]Chen, M.H. and Shao, Q.M.Monte Carlo estimation of Bayesian credible and HPD inter-vals, Journal of Computational and Graphical Statistics8, 69–92, 1999.
[6]Cohen, A.C.Progressively censored samples in life testing, Technometrics5, 327–339, 1963.
[7]Cordeiro G.M., Ortega, E.M. and Popovic, B.V.The gamma-linear failure rate distribution:theory and applications, Journal of Statistical Computation and Simulation84, 2408–2426,2014.
[8]Dempster, A.P., Laird, N.M. and Rubin, D.B.Maximum likelihood from incomplete datavia the EM algorithm, J. Roy. Statist. Soc. Ser. B39, 1–38, 1977.
[9]Efron, B. and Tibshirani, R.J.An Introduction to the Bootstrap, Chapman and Hall, NewYork, 1993.
[10]Gelfand, A.E. and Smith, A.F.M.Sampling-based approaches to calculating marginal den-sities, Journal of the American Statistical Association85, 398–409, 1990.
[11]Ghitany, M.E. and Kotz, S.Reliability properties of extended linear failure rate distributions,Probability in the Engineering and Informational Sciences21, 441–450, 2007.
[12]Jiang, H., Xie, M. and Tang, L.C.On MLEs of the parameters of a modified Weibulldistribution for progressively type-II censored samples, Journal of Applied Statistics37,617–627, 2010.
[13]Kinaci, I.Estimation ofP(Y
[14]Kus, C. and Kaya, M.F.Estimation of parameters of the loglogistic distribution based onprogressive censoring using the EM algorithm, Hacettepe Journal of Mathematics and Sta-tistics35, 203–211, 2006.
[15]Lin, C.T., Wu. S.J.S. and Balakrishnan, N.Parameter estimation for the linear hazard ratedistribution based on records and inter-record times, Communications in Statistics–Theoryand Methods32, 729–748, 2003.
[16]Lin, C.T., Wu. S.J.S. and Balakrishnan, N.Monte Carlo methods for Bayesian inferenceon the linear hazard rate distribution, Communications in Statistics–Simulation and Com-putation35, 575–590, 2006.
[17]Linhart, H. and Zucchini, W.Model Selection, Wiley, New York, 1986.
[18]Makeham, W.M.On the law of mortality and the construction of annuity tables, J. Inst.Actuar. Assur. Mag.8, 301–310, 1860.
[19]Miller, R.G.Survival Analysis, John Wiley, New York, 1981.
[20]Mugdadi, A.R.The least squares type estimation of the parameters in the power hazardfunction, Applied Mathematics and Computation169, 737–748, 2005.
[21]Mugdadi, A.R. and Min, A.Bayes estimation of the power hazard function, Journal ofInterdisciplinary Mathematics12, 675–689, 2009.
[22]Ng, T., Chan, C.S. and Balakrishnan, N.Estimation of parameters from progressively cen-sored data using EM algorithm, Computational Statistics and Data Analysis39, 371–386,2002.
[23]Pandey, A., Singh, A. and Zimmer, W.J.Bayes estimation of the linear hazard rate model,IEEE Transactions on Reliability42, 636–640, 1993.
[24]Parsian, A. and Nematollahi, N.Estimation of scale parameter under entropy loss function,Journal of Statistical Planning and Inference52, 377–391, 1996.
[25]Rastogi, M.K., Tripathi, Y.M. and Wu, S.J.Estimating the parameters of a bathtub-shapeddistribution under progressive type-II censoring, Journal of Applied Statistics39, 2389–2411,2012.
[26]Rastogi, M.K. and Tripathi, Y.M.Parameter and reliability estimation for an exponen-tiated half-logistic distribution under progressive type-II censoring, Journal of StatisticalComputation and Simulation84, 1711–1727, 2014.
[27]Rinne, H.The hazard rate: theory and inference, Justus-Liebig University Press, 2014.
[28]Sarhan, A.M. and Kundu, D.Generalized linear failure rate distribution, Communicationsin statistics: Theory and Methods38, 642–666, 2009.
[29]Sarhan, A.M. and Zaindin, M.Modified Weibull distribution, Applied Sciences11, 123–136,2009.
[30]Sen, A. and Bhattacharyya, G.Inference procedures for the linear failure rate model, Journalof Statistical Planning and Inference46, 59–76, 1995.
[31]Sen, A., Kannan, N. and Kundu, D.Bayesian planning and inference of a progressivelycensored sample from linear hazard rate distribution, Computational Statistics and DataAnalysis63, 108-121, 2013.
[32]Tian, Y., Tian, M. and ZHU, Q.A new generalized linear exponential distribution and itsapplications, Acta Mathematicae Applicatae Sinica, English Series30, 1049–1062, 2014.
[33]Tierney, L.Markov chains for exploring posterior distributions, The Annals of Statistics22, 1701–1728, 1994.
[34]Varian, H.A Bayesian approach to real estate assessment, Studies in Bayesian Econometricsand Statistics in Honor of Leonard J. Savege, Amsterdam: North Holland 195–208, 1975.