Conditional test for rank in bivariate canonical correlation analysis
Conditional test for rank in bivariate canonical correlation analysis
复制标题
双变量典型相关分析中的排名条件检验
DOI：
10.1093/biomet/88.3.874
复制
发表时间：
2001
期刊：
影响因子：
2.7
通讯作者：
B. Nielsen
中科院分区：
文献类型：
--
作者：
B. Nielsen
两个典型相关中较小者为零的假设的似然比检验是不相似的;检验统计量的分布取决于最大典型相关的值。在实际应用中，通常需要对干扰参数进行估计，本文描述了在估计量的条件下，检验的分布性质。虽然这些属性依赖于nuisance参数，但实际上这种依赖性似乎可以忽略不计。
SUMMARY The likelihood ratio test for the hypothesis that the smaller of two canonical correlations is zero is nonsimilar; the distribution of the test statistic depends on the value of the largest canonical correlation. In applications the nuisance parameter usually has to be estimated, and this paper describes the distributional properties of the test, conditional on the estimator. Although these properties depend on the nuisance parameter, the dependency seems to be negligible for practical purposes.