摘要
本文提出两个均匀性统计量用于检验多元正态性.该检验建立在多元正态分布的一个特征性质基础上.模拟研究和实例分析显示该均匀性统计量可以帮助解释来自已有正态性检验统计量的结论.
Based on a characterization for the multivariate normal distribution, we propose two uniform tests for detecting non-multinormality. Monte Carlo study and an analysis of a real data set show that the two uniform tests can help explain conclusions obtained from other existing tests of normality.
出处
《应用概率统计》
CSCD
北大核心
2005年第3期267-277,共11页
Chinese Journal of Applied Probability and Statistics
基金
This work was supported by a UNH 2003 Summer Faculty Fellowship & Research Award.
关键词
多元正态性
模拟研究
球分布
T-分布
均匀性检验
Multinormality, Monte Carlo study, spherical distribution, t-distribution, uniform test.
