摘要:

Admissibility as well as inadmissibility results are given for the case of a one-parameter exponential family. .Especially, a sufficient condition for the admissibility of a rational estimator is derived; and inadmissible estimators for the canonical parameter of i he exponential family are improved if they have a certain relatively general structure, Finally, examples of improved estimators are given

ISSN:0233-1888    

DOI:10.1080/02331888508801862

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

ISSN:0233-1888    

DOI:10.1080/02331888508801863

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

摘要:

The technique of smoothing H function in a reproducing kernel Hilbert space

ISSN:0233-1888    

DOI:10.1080/02331888508801864

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

ISSN:0233-1888    

DOI:10.1080/02331888508801865

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

摘要:

Working with the linear regression model (1.1) and having the extraneous information (1.2) about regression coefficients the problem exists how to build estimators (1.3) with the risk (1.4) which enable to utilize the known information in order to reduce their risk as compared with the risk (1.6) of the LSE (1.5). Solution of this problem is known for the positive definite matrix T, namely in form for estimators (1.8) and (1.10).First, it is shown that the proposed estimators (2.6),(2.9) and (2.16) based on psedoinversions of the matrix L represent the solution of the problem of the positive semidefinite matrix T=L'L.Further, the problem of interpretability of estimators in the sense of the inequality (3.1) exists; it is shown that all mentioned estimators are at least partially interpretable in the sense of requirements (3.2) or (3.10).

ISSN:0233-1888    

DOI:10.1080/02331888508801866

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

ISSN:0233-1888    

DOI:10.1080/02331888508801867

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

ISSN:0233-1888    

DOI:10.1080/02331888508801868

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

摘要:

Assuming that for a classification problem there are several classification procedures and it is our aim to combine these procedures in an appropriate way. We consider convex linear combinations and determine such with minimax or maximin properties. We discuss some stability properties and give examples. Furthermore we construct by

ISSN:0233-1888    

DOI:10.1080/02331888508801869

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

摘要:

In a joint paper by E. PLESZCZYKSKA and the author some equivalence relations for statistical spaces were introduced and investigated. In the present paper it is shown that those relations are preserved by products, i.e. if the corresponding factors are isomor-phie, near or almost equal statistical spaces, then their products are isomorphic, near or almost equal, respectively

ISSN:0233-1888    

DOI:10.1080/02331888508801870

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor

摘要:

Sequential estimation problems for. the exponential class of processes with

ISSN:0233-1888    

DOI:10.1080/02331888508801871

出版商:Akademie-Verlag    

年代:1985

数据来源: Taylor