EDGEWORTH EXPANSIONS FOR STATISTICS WHICH ARE FUNCTIONS OF LATTICE AND NON-LATTICE VARIABLES

被引：6

作者：

BABU, GJ

机构：

[1] Department of Statistics, Pennsylvania State University, University Park

来源：

STATISTICS & PROBABILITY LETTERS | 1991年 / 12卷 / 01期

基金：

美国国家科学基金会;

关键词：

CRAMER CONDITION; MEAN ABSOLUTE DEVIATION; LATTICE DISTRIBUTIONS; EDGEWORTH EXPANSIONS;

D O I：

10.1016/0167-7152(91)90158-N

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：

Let G be a distribution function on R(k + 1) such that the (k + 1)th marginal is lattice. Let Z(n)BAR denote the sample mean of n independent observations from G. For s greater-than-or-equal-to 3, the s-term Edgeworth expansions are obtained for a wide class of statistics which are smooth functions of Z(n)BAR. The result is then applied to a statistic similar to the Student's t-statistic, where the scaling factor, the sample standard deviation is replaced by the more robust mean absolute deviation.

引用

页码：1 / 7

页数：7

共 6 条

[1] ON EDGEWORTH EXPANSIONS IN THE MIXTURE CASES [J].