Numerical evaluation of singular multivariate normal distributions

被引：40

作者：

Genz, A

Kwong, KS

机构：

[1] Natl Univ Singapore, Dept Stat & Appl Probabil, Singapore 119260, Singapore

[2] Washington State Univ, Dept Math, Pullman, WA 99164 USA

来源：

JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION | 2000年 / 68卷 / 01期

关键词：

multivariate normal; singular distribution; numerical integration; statistical computation;

D O I：

10.1080/00949650008812053

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

We present an efficient and accurate method to evaluate multivariate normal probabilities with arbitrary singular correlation matrices. The new method is applied to the construction of simultaneous confidence intervals and simultaneous all pairwise confidence intervals for multinomial proportions when the sample size is sufficiently large.

引用

页码：1 / 21

页数：21

共 28 条

[1]

ANGELINI P, 1974, Progress in Cardiovascular Diseases, V16, P469, DOI 10.1016/0033-0620(74)90007-3

[2] LARGE SAMPLE SIMULTANEOUS CONFIDENCE-INTERVALS FOR THE MULTINOMIAL PROBABILITIES BASED ON TRANSFORMATIONS OF THE CELL FREQUENCIES [J].