Existence and strong consistency of maximum likelihood estimates for 1-dimensional exponential families

被引：1

作者：

Miao, WW

Hahn, MG

机构：

[1] MT HOLYOKE COLL,DEPT MATH,S HADLEY,MA 01075

[2] TUFTS UNIV,DEPT MATH,MEDFORD,MA 02155

来源：

STATISTICS & PROBABILITY LETTERS | 1996年 / 28卷 / 01期

基金：

美国国家科学基金会;

关键词：

maximum likelihood; strong consistency; exponential families; steep exponential families;

D O I：

10.1016/0167-7152(95)00075-5

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

The paper focuses on existence, in the sense of Hoffmann-Jorgensen, of the maximum likelihood estimate (MLE) of the parameter for 1-dimensional exponential families. It is established that the MLE exists in most cases.

引用

页码：9 / 21

页数：13

共 10 条

[1] BARNDORFFNIELSE.O, 1978, INFORMATION EXPONENT
[2] CONSISTENCY AND ASYMPTOTIC NORMALITY OF MLES FOR EXPONENTIAL MODELS
BERK, RH
[J]. ANNALS OF MATHEMATICAL STATISTICS, 1972, 43 (01): : 193 - &
[3] BROWN LD, 1986, IMS MONOGRAPH
[4] GEOMETRY OF EXPONENTIAL FAMILIES
EFRON, B
[J]. ANNALS OF STATISTICS, 1978, 6 (02) : 362 - 376
[5] GEYER CJ, 1990, THESIS U WASHINGTON
[6] HOFFMANNJORGENS.J, 1994, PROBABILITY VIEW STA, V2
[7] MIAO W, 1996, UNPUB EXISTENCE STRO
[8] Rockafellar R., 1972, CONVEX ANAL
[9] NOTE ON THE CONSISTENCY OF THE MAXIMUM LIKELIHOOD ESTIMATE
WALD, A
[J]. ANNALS OF MATHEMATICAL STATISTICS, 1949, 20 (04): : 595 - 601
[10] [No title captured]

← 1 →