Kybernetika 47 no. 4, 595-611, 2011

An admissible minimax estimator of a lower-bounded scale parameter under squared-log error loss function

Eisa Mahmoudi and Hojatollah Zakerzadeh

Abstract:

Estimation in truncated parameter space is one of the most important features in statistical inference, because the frequently used criterion of unbiasedness is useless, since no unbiased estimator exists in general. So, other optimally criteria such as admissibility and minimaxity have to be looked for among others. In this paper we consider a subclass of the exponential families of distributions. Bayes estimator of a lower-bounded scale parameter, under the squared-log error loss function with a sequence of boundary supported priors is obtained. An admissible estimator of a lower-bounded scale parameter, which is the limiting Bayes estimator, is given. Also another class of estimators of a lower-bounded scale parameter, which is called the truncated linear estimators, is considered and several interesting properties of the estimators in this class are studied. Some comparisons of the estimators in this class with an admissible estimator of a lower-bounded scale parameter are presented.

Keywords:

admissibility, Bayes estimator, truncated parameter spaces, squared-log error loss

Classification:

62C10, 62C15, 62C20

paper.pdf

References:

  1. J. C. Berry: Minimax estimation of a restricted exponential location parameter. Statist. Decision 11 (1993), 307-316.   CrossRef
  2. C. R. Blyth: Minimax statistical procesures and their admissibility. Ann. Math. Statist. 22 (1951), 22-42.   CrossRef
  3. L. Brown: Inadmissibility of the usual estimators of scale parameters in problems with uknown location and scale parameters. Ann. Math. Statist. 29(1) (1968), 29-48.   CrossRef
  4. T. S. Ferguson: Mathematical Statistics: A Decision Theoretic Approach. Academic Press, New York 1967.   CrossRef
  5. D. C. Hoaglin: The small-sample variance of the Pitman location estimators. J. Amer. Statist. Assoc. 70 (1975), 880-888.   CrossRef
  6. M. Jafari Jozani, N. Nematollahi and K. Shafie: An admissible minimax estimator of a bounded scale-parameter in a subclass of the exponential family under scale-invariant squared-error loss. Statist. Prob. Letter {\mi60} (2002), 434-444.   CrossRef
  7. W. Katz: Admissible and minimax estimator of parameters in truncated space. Ann. Math. Statist. {\mi 32} (1961), 136-142.   CrossRef
  8. E. L. Lehmann and G. Casella: Theory of Point Estimation. Second edition. Springer-Verlag, John Wiley, New York 1998.   CrossRef
  9. J. J. A. Moors: Estimation in Truncated Parameter Spaces. Ph.D Thesis, Tilburg University Tilburg, The Netherlands 1985.   CrossRef
  10. J. J. A. Moors and J. C. van Houwelingen: Estimation of linear models with inequality restrictions. Statist. Neerlandica 47 (1993), 185-198.   CrossRef
  11. A. Parsian and N. Nematollahi: Estimation of scale parameter under entropy loss function. J. Statis. Plann. Infer. 52 (1996), 77-91.   CrossRef
  12. E. J. J. Pitman: The estimation of location and scale parameters of a continuous population of any given form. Biometrika {\mi 30} (1938), 391-421.   CrossRef
  13. E. J. J. Pitman: Some Basic Theory for Statistical Inference. Chapman Hall, London 1979.   CrossRef
  14. M. S. Rahman and R. P. Gupta: Family of transformed chi-square distributions. Comm. Statist. Theory Methods 22 (1993), 135-146.   CrossRef
  15. T. Robertson, F. T. Wright and R. L. Dijkstra: Order Restricted Statistical Inference. John Wiley, New York 1988.   CrossRef
  16. N. Sanjari Farsipour and H. Zakerzadeh: Estimation of a gamma scale parameter under asymmetric squared-log error loss. Comm. Statist. Theory Methods 34 (2005), 1-9.   CrossRef
  17. P. Shao and W. E. Strawderman: Improving on truncated linear estimates of exponential and gamma scale parameters. Canad. J. Statist. 24 (1996), 105-114.   CrossRef
  18. C. Stein: The admissibility of Pitman's estimator for a single location parameter. Ann. Math. Statist. 30 (1959), 970-979.   CrossRef
  19. C. van Eeden: Minimax estimation of am lower-bounded scale parameter of a gamma distribution for scale invariant squared-error loss. Canada. J. Statist. {\mi 23} (1995), 245-256.   CrossRef
  20. C. van Eeden: Minimax estimation of a lower-bounded scale-parameter of an F-distribution. Statist. Prob. Lett. 46 (2000), 283-286.   CrossRef
  21. C. van Eeden and J. V. Zidek: Group-Bayes estimation of the exponential mean: A retrospective view of the wald theory. In: Statistical Decision Theory and Related Topics, V (S. S. Gupta and J. Berger, eds.), Springer, Berlin 1994, pp. 35-49.   CrossRef
  22. C. van Eeden and J. V. Zidek: Group-Bayes estimation of the exponential mean: A preposterior analysis. Test 3 (1994), 125-143.   CrossRef
  23. C. van Eeden and J. V. Zidek: Correction to Group-Bayes estimation of the exponential mean: A preposterior analysis. Test 3 (1994), 247.   CrossRef