Saddlepoint approximations for distributions of quadratic forms in normal variables

被引：111
作者：
Kuonen, D [1 ]
机构：
[1] Swiss Fed Inst Technol, Dept Math, CH-1015 Lausanne, Switzerland
来源：
BIOMETRIKA | 1999年 / 86卷 / 04期
关键词：
chi-squared variable; nonparametric regression; normal variable; pseudolikelihood ratio test; quadratic form; saddlepoint approximation;
D O I：
10.1093/biomet/86.4.929
中图分类号：
Q [生物科学];
学科分类号：
07 ; 0710 ; 09 ;
摘要：
An extensive study and several applications show that a saddlepoint approximation for the distribution of quadratic forms in normal variates works very well and outperforms existing approximate methods in accuracy.
引用
收藏
页码：929 / 935
页数：7
相关论文
共 15 条
[1]  
AZZALINI A, 1989, BIOMETRIKA, V76, P1
[2]  
BARNDORFFNIELSEN OE, 1990, J ROY STAT SOC B MET, V52, P485
[3]  
Bowman A.W., 1997, Applied Smoothing Techniques for Data Analysis: the Kernel Approach with S-Plus Illustrations, VVolume 18
[4]   SADDLEPOINT APPROXIMATIONS IN STATISTICS [J].
DANIELS, HE .
ANNALS OF MATHEMATICAL STATISTICS, 1954, 25 (04) :631-650
[5]  
Davies R. B., 1980, Journal of the Royal Statistical Society: Series C, V29, P323, DOI DOI 10.2307/2346911
[6]  
FAREBROTHER RW, 1990, APPL STAT-J ROY ST C, V39, P294
[7]   COMPUTING DISTRIBUTION OF QUADRATIC FORMS IN NORMAL VARIABLES [J].
IMHOF, JP .
BIOMETRIKA, 1961, 48 (3-4) :419-&
[8]  
Jensen J.L., 1995, SADDLEPOINT APPROXIM
[9]  
Johnson N. L., 1970, DISTRIBUTIONS STAT C, V2
[10]  
KOERTS J, 1969, THEORY APPL GEN LINE
12