Estimating functions of canonical correlation coefficients (Q1069618)

Let \(\rho^ 2_ 1,...,\rho^ 2_ p\) be the squares of the population canonical correlation coefficients from a normal distribution. This paper is concerned with the estimation of the parameters \(\delta_ 1,...,\delta_ p\), where \(\delta_ i=\rho^ 2_ i/(1-\rho^ 2_ i)\), \(i=1,...,p\), in a decision theoretic way. The approach taken is to estimate a parameter matrix \(\Delta\) whose eigenvalues are \(\delta_ 1,...,\delta_ p\), given a random matrix F whose eigenvalues have the same distribution as \(r^ 2_ i/(1-r^ 2_ i)\), \(i=1,...,p\), where \(r_ 1,...,r_ p\) are the sample canonical correlation coefficients.