Non-parametric randomness tests based on success runs of fixed length

被引：26

作者：

Koutras, MV ^{[1
]}

Alexandrou, VA ^{[1
]}

机构：

[1] UNIV ATHENS,DEPT STAT,ATHENS 15784,GREECE

来源：

STATISTICS & PROBABILITY LETTERS | 1997年 / 32卷 / 04期

关键词：

Bernoulli trials; success runs; non-parametric tests of randomness; Markov chain imbeddable variables; generating functions;

D O I：

10.1016/S0167-7152(96)00099-5

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：

Let X(n) be a random variable enumerating the number of appearances of a specific pattern in a sequence of n Bernoulli trials. A new method is presented for obtaining the conditional distribution of X(n) given the number of successes in the n trials. The method is applied to three fixed-length run statistics and the results are used for establishing and investigating certain non-parametric tests of randomness.

引用

页码：393 / 404

页数：12

共 20 条

[1]

DYCK PJ, 1984, PERIPHERAL NEUROPATH, V1, P1103

[2]

Feller W., 1968, INTRO PROBABILITY TH

[3]

FU JC, 1995, ANN I STAT MATH, V47, P435

[4] DISTRIBUTION-THEORY OF RUNS - A MARKOV-CHAIN APPROACH [J].