Approximations for two-dimensional discrete scan statistics in some block-factor type dependent models

被引:4
|
作者
Amarioarei, Alexandru [1 ,2 ,3 ]
Preda, Cristian [1 ,2 ]
机构
[1] Univ Sci & Technol Lille 1, Lab Math Paul Painleve, UMR 8524, Lille, France
[2] INRIA Nord Europe Modal, Paris, France
[3] Natl Inst R&D Biol Sci, Bucharest, Romania
关键词
Scan statistics; m-dependent stationary sequences; Block-factor; STATIONARY-SEQUENCES; P-VALUES;
D O I
10.1016/j.jspi.2014.05.002
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
We consider the two-dimensional discrete scan statistic generated by a block-factor type model. Under the null hypothesis, in which the block-factor model is obtained from an i.i.d. sequence of random variables, we present an approximation for the distribution of the scan statistics and the corresponding error bounds. Numerical results are presented to evaluate the accuracy of the approximation discussed in this paper. (C) 2014 Elsevier B.V. All rights reserved.
引用
收藏
页码:107 / 120
页数:14
相关论文
共 50 条