Estimating the Mean of a Normal Distribution with Loss Equal to Squared Error Plus Complexity Cost
Kempthorne, Peter J.
Ann. Statist., Tome 15 (1987) no. 1, p. 1389-1400 / Harvested from Project Euclid
Estimating the mean of a $p$-variate normal distribution is considered when the loss is squared error plus a complexity cost. The complexity of estimates is defined using a partition of the parameter space into sets corresponding to models of different complexity. The model implied by the use of an estimate determines the estimate's complexity cost. Complete classes of estimators are developed which consist of preliminary-test estimators. As is the case when loss is just squared error, the maximum-likelihood estimator is minimax. However, unlike the no-complexity-cost case, the maximum-likelihood estimator is inadmissible even in the case when $p = 1$ or 2.
Publié le : 1987-12-14
Classification:  Admissibility,  complete class,  generalized Bayes,  minimax,  preliminary-test estimators,  62C07,  62C20
@article{1176350600,
     author = {Kempthorne, Peter J.},
     title = {Estimating the Mean of a Normal Distribution with Loss Equal to Squared Error Plus Complexity Cost},
     journal = {Ann. Statist.},
     volume = {15},
     number = {1},
     year = {1987},
     pages = { 1389-1400},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176350600}
}
Kempthorne, Peter J. Estimating the Mean of a Normal Distribution with Loss Equal to Squared Error Plus Complexity Cost. Ann. Statist., Tome 15 (1987) no. 1, pp.  1389-1400. http://gdmltest.u-ga.fr/item/1176350600/