
Get the probability that for the selected CCA object the chosen canonical correlation coefficient is different from zero.
Wilks' statistic: the probability that coefficient ρ_{index} differs from zero is
probability = chiSquareQ (χ^{2}, ndf), 
where the number of degrees of freedom parameter equals
ndf = (n_{y}  index +1)(n_{x}  index +1) 
and the chisquared parameter is
χ^{2} = –(numberOfObservations  (n_{y} + n_{x} +3)/2) log (Λ_{index}), 
In the formulas above the variables n_{y} and n_{x} are the dimensions of the dependent and the independent data sets whose canonical correlations have been obtained, and Wilks' lambda is:
Λ_{index} = Π_{i=index..min(ny,nx)} (1 – ρ_{i}^{2}) 
© djmw, April 7, 2004