Advertisement

Likelihood Testing With Censored and Missing Duration Data

  • Narayanaswamy Balakrishnan
  • Milan Stehlík
Article

Abstract

Duration data of a component, product, or system are often incomplete. We give guidelines for likelihood ratio testing for small samples with some data imperfections, that is, when some information is missing and some information is censored. We introduce the model of the missing time-to-failure mechanism, which still enables the exact likelihood ratio testing of the scale and homogeneity. Such a model encompasses well both the generalized gamma and Pareto duration time models with missing individual times to failure. The exact distribution of the likelihood ratio test for the censored sample from Weibull distribution with known shape parameter is derived and discussed. We discuss separately Type I, Type II, and progressively censored samples. We show that the Type I censoring and Type II censoring differ substantially. The construction of the pivotal quantity is possible for Type II and progressively Type II censoring; however, it is not available for the case of Type I censoring. Thus, for the latter case the usage of the exact likelihood ratio test is a natural option. We also discuss the case of the generalized gamma distribution. For the case with nuisance shape parameters we use an integrated likelihood approach. Convenient examples illustrate the methods developed in the article.

Keywords

Exact test Integrated likelihood Lambert W function Likelihood ratio Missing data Nuisance parameter Profile likelihood Progressive censoring Type I censoring Type II censoring 

AMS Subject Classification

62E15 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Amstadter, B. L. 1971. Reliability mathematics, chap. 9. New York, NY: McGraw-Hill.Google Scholar
  2. Alvarez-Andradea, S., N. Balakrishnan, and L. Bordes. 2007. Homogeneity tests based on several progressively Type-II censored samples. J. Multivariate Anal., 98, 1195–1213.MathSciNetCrossRefGoogle Scholar
  3. Bahadur, R. R. 1965. An optimal property of the likelihood ratio statistic. In Proc. 5th Berkeley Symposium on Probability Theory and Mathematical Statistics, vol. 1, ed. L. Le Cam and J. Neyman, 13–26. Berkeley, CA: University of California Press.Google Scholar
  4. Bain, L. J., and M. Engelhardt. 1992. Introduction to probability and mathematical statistics (2nd ed.). Boston, MA: PWSKENT Publishing Company.Google Scholar
  5. Balakrishnan, N. 2007. Progressive censoring methodology: An appraisal. TEST, 16, 211–259.MathSciNetCrossRefGoogle Scholar
  6. Balakrishnan, N., and R. Aggarwala. 2000. Progressive censoring theory, methods, and applications, Series: Statistics for Industry and Technology, XV. Boston, MA: Birkhäuser.CrossRefGoogle Scholar
  7. Balakrishnan, N., and R. A. Sandhu. 1996. Best linear unbiased and maximum likelihood estimation for exponential distributions under general progressive Type-II censored samples. Sankhya, Ser. B, 58, 1–9.MathSciNetzbMATHGoogle Scholar
  8. Barndorf-Nielsen, O. E. 1978. Information and exponential families in statistical theory, 111–115. New York, NY: John Wiley Sons.Google Scholar
  9. Bartholomew, D. J. 1963. The sampling distribution of an estimate arising in life testing. Technometrics, 5, 361–374.MathSciNetCrossRefGoogle Scholar
  10. Childs, A., B. Chandrasekar, N. Balakrishnan, and D. Kundu. 2003. Exact likelihood inference based on Type I and Type II hybrid censored samples from the exponential distribution. Ann. Inst. Stat. Math., 55(2), 319–330.zbMATHGoogle Scholar
  11. Ciuperca, G. 2002. Likelihood ratio statistic for exponential mixtures. Ann. Inst. Stat. Math. 54(3), 585–594.MathSciNetCrossRefGoogle Scholar
  12. Cohen, A. C. 1991. Truncated and censored samples. Statistics, a Series of Textbooks and Monographs. New York, NY: Marcel Dekker.CrossRefGoogle Scholar
  13. Coit, D. W., and T. Jin. 2000. Gamma distribution parameter estimation for field reliability data with missing failure times. IIE Trans. 32, 1161–1166.Google Scholar
  14. Coit, D. W., and Dey, K. A. 1999. Analysis of grouped data from field-failure reporting systems. Reliability Eng. System Safety, 65, 95–101.CrossRefGoogle Scholar
  15. Cole, K. N., P. B. Nagarsenker, and B. N. Nagarsenker. 1987. A test for equality of exponential distributions based on Type-II censored samples. IEEE Trans. Reliability, 36(1), 94–97.CrossRefGoogle Scholar
  16. Corless, R. M., G. H. Gonnet, D. E. G. Hare, D. J. Jeffrey, and D. E. Knuth. 1996. On the Lambert W function. Adv. Comput. Math. 5, 329–359.MathSciNetCrossRefGoogle Scholar
  17. Corless, R. M., G. H. Gonnet, D. E. G. Hare, D. J. Jeffrey and D. E. Knuth. 1993. Lambert’s W function in Maple. Maple Tech. Newslett., 9, 12–22.Google Scholar
  18. Dey, K. A. 1982. Statistical analysis of noisy and incomplete failure data. In Proceedings Annual Reliability and Maintainability Symposium (RAMS), IEEE, Piscataway, NJ, 93–97.Google Scholar
  19. Economou, P., and M. Stehlík. 2014. On small samples testing for frailty through homogeneity test. Commun. Stat. Simul. Comput. http://dx.doi.org/10.1080/03610918.2013.763982.
  20. Epstein, B., and M. Sobel. 1954. Some theorems relevant to life testing from an exponential distribution. Ann. Math. Stat. 25, 373–381.MathSciNetCrossRefGoogle Scholar
  21. Gaudoin, O., and J. L. Soler. 1992. Statistical analysis of the geometric de-eutrophication software reliability model. IEEE Trans. on Reliability, 41(4), 518–524.CrossRefGoogle Scholar
  22. Gautam, N. 1999. Erlang distribution: Description and case study. In Industrial engineering applications and practice: Users encyclopedia, eds. A. Mitaland J. Chen.Google Scholar
  23. Hamada, M. S., A. G Wilson, C. S. Reese, and H. F. Martz. 2008. Bayesian reliability. Springer Series in Statistics. New York, NY: Springer.CrossRefGoogle Scholar
  24. Johansen, S. 1979. Introduction to the theory of regular exponential families. Lecture Notes, vol. 3. Copenhagen, Denmark: Institute of Mathematical Statistics, University of Copenhagen.Google Scholar
  25. Kleinrock, L. 1975. Queueing systems, vol. 1, 71–72, 119–134. Toronto, Canada: John Wiley & Sons.zbMATHGoogle Scholar
  26. Kundu, D., and A. Joarder. 2006. Analysis of Type-II progressively hybrid censored data. Comput. Stat. Data Anal. 50, 2509–2528.MathSciNetCrossRefGoogle Scholar
  27. Lehmann, E. L., and J. P. Romano. 2005. Testing statistical hypotheses. New York, NY: Springer-Verlag, LLC.zbMATHGoogle Scholar
  28. Lin, D. K. J., J. S. Usher, and F. M. Guess. 1996. Bayes estimation of component from masked system-life data. IEEE Trans. Reliability, 45, 233–237.CrossRefGoogle Scholar
  29. Little, R. J. A., and D. B. Rubin. 1987. Statistical analysis with missing data. New York, NY: Wiley.zbMATHGoogle Scholar
  30. Lomax, K. S. 1954. Business failures: Another example of the analysis of failure data. J. Am. Stat. Assoc. 49, 847–852.CrossRefGoogle Scholar
  31. Marshall, A. W., and I. Olkin. 2007. Life distributions. New York, NY: Springer.zbMATHGoogle Scholar
  32. Mosler, K., and L. Haferkamp. 2007. Size and power of recent tests for homogeneity in exponential mixtures. Commun. Stat. Simulation Comput. 36, 493–504.MathSciNetCrossRefGoogle Scholar
  33. Mosler, K., and C. Scheicher. 2008. Homogeneity testing in a Weibull mixture model. Stat. Papers, 49, 315–332.MathSciNetCrossRefGoogle Scholar
  34. Nagarsenker, P. B. 1980. On a test of equality of several exponential survival distributions. Biometrika, 67(2), 475–478.MathSciNetCrossRefGoogle Scholar
  35. Philbrick, S. W. 1985. A practical guide to the single parameter Pareto distribution. Proc. Casualty Actuarial Society, LXXII, 44–84.Google Scholar
  36. Rublík, F. 1989a. On optimality of the LR tests in the sense of exact slopes, Part 1, General case. Kybernetika, 25, 13–25.MathSciNetzbMATHGoogle Scholar
  37. Rublík, F. 1989b. On optimality of the LR tests in the sense of exact slopes, Part 2, Application to individual distributions. Kybernetika, 25, 117–135.MathSciNetzbMATHGoogle Scholar
  38. Soland, R. M. 1966. Use of Weibull distribution in Bayesian decision theory, Report No. RAC-TP-225. McLean, VA: Research Analysis Corporation.Google Scholar
  39. Soland, R. M. 1969. Bayesian analysis of the Weibull process with unknown scale and shape parameters. IEEE Trans. Reliability, 18(4), 181–184.CrossRefGoogle Scholar
  40. Severini, T. A. 1999. On the relationship between Bayesian and non-Bayesian elimination of nuisance parameters. Stat. Sin. 9, 713–724.MathSciNetzbMATHGoogle Scholar
  41. Severini, T. A. 2006. Likelihood methods in statistics. Oxford Statistical Science Series. New York, NY: Oxford University Press.zbMATHGoogle Scholar
  42. Severini, T. A. 2010. Likelihood ratio statistics based on an integrated likelihood. Biometrika, 97(2), 481–496.MathSciNetCrossRefGoogle Scholar
  43. Shoukri, M. M. 1987. Simple Bayes test of equality of exponential means. IEEE Trans. Reliability, 36(5), 613–616.CrossRefGoogle Scholar
  44. Stehlík, M. 2003. Distributions of exact tests in the exponential family. Metrika, 57, 145–164.MathSciNetCrossRefGoogle Scholar
  45. Stehlík, M. 2006. Exact likelihood ratio scale and homogeneity testing of some loss processes. Stat. Probability Lett., 76, 19–26.MathSciNetCrossRefGoogle Scholar
  46. Stehlík, M. 2007. Exact testing of the scale with the missing time-to-failure information. Commun. Dependability Qual. Manage. Eng. (CDQM), 10(2), 124–129.Google Scholar
  47. Stehlík, M. 2008. Homogeneity and scale testing of generalized gamma distribution. Reliability Eng. System Safety, 93, 1809–1813.CrossRefGoogle Scholar
  48. Stehlík, M. 2009. Scale testing in small samples with missing time-to-failure information. Int. J. Reliability Qual. Safety Eng. (IJRQSE), 16(6), 1–13.CrossRefGoogle Scholar
  49. Stehlík, M., R. Potocký, H. Waldl, and Z. Fabián. 2010. On the favourable estimation of fitting heavy tailed data. Comput. Stat., 25, 485–503.CrossRefGoogle Scholar
  50. Stehlík, M., and H. Wagner. 2011. Exact likelihood ratio testing for homogeneity of exponential distribution. Commun. Stat. Simulation Comput., 40, 663–684.MathSciNetCrossRefGoogle Scholar
  51. Stehlík, M., P. Economou, J. Kiseľák, and W.-D. Richter. 2014. Kullback-Leibler life time testing. App. Math. Comput., 240, 122–139.MathSciNetCrossRefGoogle Scholar
  52. Sukhatme, P. V. 1937. Tests of significance for samples of the χ2 population with two degrees of freedom. Ann. Eugenics, 8, 52–56.CrossRefGoogle Scholar
  53. Thiagarajah, K. R. 1995. Homogeneity tests for scale parameters of 2-parameter exponential populations under time censoring. IEEE Trans. Reliability. 44(2), 297–301.CrossRefGoogle Scholar
  54. Thomas, D. R., and W. M. Wilson. 1972. Linear order statistic estimation for the two parameter Weibull and extreme value distributions from Type-II progressively censored samples. Technometrics, 14, 679–691.CrossRefGoogle Scholar
  55. Usher, J. S. 1996. Weibull component reliability—Prediction in the presence of masked data. IEEE Trans. Reliability, 45, 229–232.CrossRefGoogle Scholar
  56. Viveros, R., and N. Balakrishnan. 1994. Interval estimation of parameters of life from progressively censored data. Technometrics, 36(1), 84–91.MathSciNetCrossRefGoogle Scholar
  57. Wilks, S. S. 1962. Mathematical statistics. New York, NY: John Wiley & Sons.zbMATHGoogle Scholar
  58. Wu, S. J., D. H. Chen, and S. T. Chen. 2006. Bayesian inference for Rayleigh distribution under progressive censored sample. App. Stochastic Models Business Ind., 22, 269–279.MathSciNetCrossRefGoogle Scholar

Copyright information

© Grace Scientific Publishing 2015

Authors and Affiliations

  1. 1.Department of Mathematics and StatisticsMcMaster UniversityHamiltonCanada
  2. 2.Department of Applied StatisticsJohannes Kepler UniversityLinzAustria
  3. 3.Departamento de MatemáticaUniversidad Técnica Federico Santa MaríaValparaísoChile

Personalised recommendations