A unified approach to non-minimaxity of sets of linear combinations of restricted location estimators

Tatsuya Kubokawa, William E. Strawderman

Research output: Contribution to journalArticlepeer-review


This paper studies minimaxity of estimators of a set of linear combinations of location parameters μi, i=1,...,k under quadratic loss. When each location parameter is known to be positive, previous results about minimaxity or non-minimaxity are extended from the case of estimating a single linear combination, to estimating any number of linear combinations. Necessary and/or sufficient conditions for minimaxity of general estimators are derived. Particular attention is paid to the generalized Bayes estimator with respect to the uniform distribution and to the truncated version of the unbiased estimator (which is the maximum likelihood estimator for symmetric unimodal distributions). A necessary and sufficient condition for minimaxity of the uniform prior generalized Bayes estimator is particularly simple. If one estimates θ=Atμ where A is a k×ℓ known matrix, the estimator is minimax if and only if (AAt)ij≤0 for any i and j (i≠j). This condition is also sufficient (but not necessary) for minimaxity of the MLE.

Original languageEnglish (US)
Pages (from-to)1429-1444
Number of pages16
JournalJournal of Multivariate Analysis
Issue number10
StatePublished - Nov 2011

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty


  • Decision theory
  • Generalized bayes
  • Linear combination
  • Location parameter
  • Location-scale family
  • Maximum likelihood estimator
  • Minimaxity
  • Primary
  • Quadratic loss
  • Restricted estimator
  • Restricted parameter
  • Secondary
  • Truncated estimator


Dive into the research topics of 'A unified approach to non-minimaxity of sets of linear combinations of restricted location estimators'. Together they form a unique fingerprint.

Cite this