摘要
研究典型相关分析的原理、典型成分的计算方法及计算步骤.把两组变量X与Y转化为具有最大相关性的若干对典型成分,直到两组变量的相关性被分解.通过典型相关系数及其显著性检验,选择典型成分分析两组变量的相关性.实例表明只有第一个典型相关系数能通过显著性检验,而其它两个典型相关系数显著为零,故应选取第一对典型成分F1和G1做分析.
Canonical correlation analysis is a statistical analysis method to study the correlation between two sets of variable. This paper introduces the principle of canonical correlation analysis, algorithm and computing steps of canonical components. The two variable sets X and Y are transformed into pairs of canonical components with the greatest number of correlations so that correlation of the two sets is decomposed. Applying the significance test to the canonical correlations, canonical components are selected to analyze the correlation of the two variable sets. An example shows that only the first canonical correlation coefficient passes the test of significance, while significances of the other canonical correlation coefficients are zero.
出处
《高等数学研究》
2011年第1期75-76,共2页
Studies in College Mathematics
关键词
典型相关分析
典型相关系数
典型成分
canonical correlation analysis, correlation coefficient, canonical components
