The Annals of Statistics

Improving on the MLE of a bounded normal mean

Éric Marchand and François Perron

Full-text: Open access


We consider the problem of estimating the mean of a $p$-variate normal distribution with identity covariance matrix when the mean lies in a ball of radius $m$. It follows from general theory that dominating estimators of the maximum likelihood estimator always exist when the loss is squared error. We provide and describe explicit classes of improvements for all problems $(m, p)$. We show that,for small enough $m$, a wide class of estimators, including all Bayes estimators with respect to orthogonally invariant priors, dominate the maximum likelihood estimator. When $m$ is not so small, we establish general sufficient conditions for dominance over the maximum likelihood estimator. These include, when $m \le \sqrt{p}$, the Bayes estimator with respect to a uniform prior on the boundary of the parameter space. We also study the resulting Bayes estimators for orthogonally invariant priors and obtain conditions of dominance involving the choice of the prior. Finally, these Bayesian dominance results are further discussed and illustrated with examples, which include (1) the Bayes estimator for a uniform prior on the whole parameter space and (2) a new Bayes estimator derived from an exponential family of priors.

Article information

Ann. Statist., Volume 29, Number 4 (2001), 1078-1093.

First available in Project Euclid: 14 February 2002

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62F10: Point estimation 62F15: Bayesian inference 62F30: Inference under constraints

Maximum likelihood estimator restricted parameter space multivariate normal distribution Bayes estimators squared error loss noncentral chi-square distribution Langevin distribution monotone likelihood ratio modified Bessel function


Marchand, Éric; Perron, François. Improving on the MLE of a bounded normal mean. Ann. Statist. 29 (2001), no. 4, 1078--1093. doi:10.1214/aos/1013699994.

Export citation


  • Berry, C. (1990). Minimax estimation of a bounded normal mean vector. J.Multivariate Anal. 35 130-139.
  • Casella, G. and Strawderman, W. E. (1981). Estimating a bounded normal mean. Ann.Statist. 9 870-878.
  • Charras, A. and van Eeden, C. (1991). Bayes and admissibility properties of estimators in truncated parameter spaces. Canad.J.Statist.19 121-134.
  • DasGupta, A. (1985). Bayes minimax estimation in multiparameter families when the parameter space is restricted to a bounded convex set. Sankhy¯a Ser.A 47 326-332.
  • Gatsonis, C., MacGibbon, B. and Strawderman, W. E. (1987). On the estimation of a restricted normal mean. Statist.Probab.Lett.6 21-30.
  • Kempthorne, P. J. (1988). Dominating inadmissible procedures using compromise decision theory. In Statistical Decision Theory and Related Topics IV (S. Gupta and J. O. Berger, eds.) 1 381-396. Springer, New York.
  • Marchand, ´E. and MacGibbon, B. (2000). Minimax estimation of a constrained binomial proportion. Statist.Decisions 18 129-167.
  • Marchand, ´E. and Perron, F. (1999). Improving on the mle of a bounded normal mean. Technical Report 2628, Centre de recherches math´ematiques, Univ. Montr´eal.
  • Moors, J. J. A. (1981). Inadmissibility of linearly invariant estimators in the truncated parameter spaces. J.Amer.Statist.Assoc.76 910-915.
  • Moors, J. J. A. (1985). Estimation in truncated parameter spaces. Ph.D. thesis, Tilburg Univ.
  • Robert, C. (1990). Modified Bessel functions and their applications in probability and statistics. Statist.Probab.Lett.9 155-161.
  • Sacks, J. (1963). Generalized Bayes solutions in estimation problems. Ann.Math.Statist.34 751-768.
  • Vidakovic, B. and DasGupta, A. (1996). Efficiency of linear rules for estimating a bounded normal mean. Sankhy¯a Ser.A 58 81-100.
  • Watson, G. S. (1983). Statistics on Spheres. Wiley, New York.