Estimation problems for the two-parameter negative binomial distribution

被引：3

作者：

Wang, YN ^{[1
]}

机构：

[1] BOWLING GREEN STATE UNIV,DEPT MATH & STAT,BOWLING GREEN,OH 43402

来源：

STATISTICS & PROBABILITY LETTERS | 1996年 / 26卷 / 02期

关键词：

negative binomial distribution; maximum likelihood estimation; unbiasedness;

D O I：

10.1016/0167-7152(94)00259-2

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：

This note shows a major problem in the proof of Anscombe's conjecture given in Aragon et al. (1992). It also proves that the unbiased estimator of the parameter k does not exist.

引用

页码：113 / 114

页数：2

共 4 条

[1] ANSCOMBE FJ, 1950, BIOMETRIKA, V37, P358, DOI 10.1093/biomet/37.3-4.358
[2] EXISTENCE AND UNIQUENESS OF THE MAXIMUM-LIKELIHOOD ESTIMATOR FOR THE 2-PARAMETER NEGATIVE BINOMIAL-DISTRIBUTION
ARAGON, J
EBERLY, D
EBERLY, S
[J]. STATISTICS & PROBABILITY LETTERS, 1992, 15 (05) : 375 - 379
[3] COMPOUND MULTINOMIAL LIKELIHOOD FUNCTIONS ARE UNIMODAL - PROOF OF A CONJECTURE OF GOOD,IJ
LEVIN, B
REEDS, J
[J]. ANNALS OF STATISTICS, 1977, 5 (01) : 79 - 87
[4] Wilson L.J., 1986, Metrika, V33, P349

← 1 →