Download e-book for iPad: A festschrift for Herman Rubin by Dasgupta A. (ed.)

By Dasgupta A. (ed.)

ISBN-10: 0940600617

ISBN-13: 9780940600614

Show description

Read or Download A festschrift for Herman Rubin PDF

Similar reference books

Download e-book for iPad: Social Security Handbook 2013: Overview of Social Security by Social Security Administration

The Bernan Press Social safety guide: assessment of Social protection courses, 2013 offers information regarding Social safety courses and providers, and identifies rights and tasks lower than the Social safeguard legislation. The guide additionally comprises information regarding comparable courses administered by means of organisations except the Social protection management.

Download PDF by Nick Maskell, Ann Millar: Oxford Desk Reference: Respiratory Medicine

Within the period of evidence-based drugs, clinicians draw upon an unlimited source of research-based facts to steer their perform. notwithstanding, discovering this knowledge isn't really consistently effortless, relatively whilst it's so much wanted. with regards to breathing drugs which covers greater than forty person stipulations, the research-based facts could be constrained and/or disproportionate to the prevalence of the .

Extra resources for A festschrift for Herman Rubin

Sample text

1958). Admissibility for estimation with quadratic loss. Ann. Math. Statist. 29, 406–436. MR124101 [19] Lai, W. (1996). Admissibility and the recurrence of Markov chains with applications. D. Thesis, University of Minnesota. [20] Lamperti, J. (1960). Criteria for the recurrence or transience of stochastic processes I. Journal of Math. Anal. Appl. 1, 314–330. MR126872 [21] Meyn, S. P. and Tweedie, R. L. (1993). Markov Chains and Stochastic Stability. New York: Springer-Verlag. MR1287609 [22] Revuz, D.

In this case Casella and Strawderman (1981) and Zinzius (1981) showed that the unique minimax estimator of the mean θ for a restriction of the form θ ∈ [−m, m] is the Bayes estimator corresponding to a 2 point prior on {−m, m} for m sufficiently small. The material in section 6 deals with this result, and the large related literature that has followed. In many problems, as in the previous paragraph, Bayes or Generalized Bayes estimators are known to form a complete class. When loss is quadratic and the prior (and hence typically the posterior) distribution is not degenerate at a point, the Bayes estimator cannot take values on the boundary of the parameter space.

K; with θ1 ≤ θ2 ≤ . . ≤ θk , Sackrowitz and Strawderman (1974) investigated the admissibility (for various loss functions) of the MLE of (θ1 , . . , θk ), while Sackrowitz (1982) provided improvements (under sum of squared error losses) to the MLE in the cases above where it is inadmissible. Further examples consist of a series of papers by Shao and Strawderman (1994,1996a,1996b) where, in various models, improvements under squared error loss to truncated estimators are obtained. Further related historical developments are given in the review paper of van Eeden (1996).

Download PDF sample

A festschrift for Herman Rubin by Dasgupta A. (ed.)


by Steven
4.2

Rated 4.66 of 5 – based on 12 votes