ccorPvals {CMO} | R Documentation |
Calculates p-values for canonical correlation values
ccorPvals(ccor, N, onlyPs = TRUE)
ccor |
A List object as derived by the function cancor. |
N |
number of independent observations the canonical correlation has been derived from |
onlyPs |
TRUE limits the output to a vector of p-values |
Canonical correlation analysis is a way of making sense of cross-covariance matrices between two sets of random variables. The canonical correlation values can be converted into a test statistic of Chi-square type to derive a significance measurement (p-value) foreach canonical correlation value.
cor |
correlations derived by the function cancor |
lambda |
cumulative product of the noise vector of the correlation vector |
chisq |
Chi-square value foreach correlation |
df |
degrees of freedom of the Chi-square distribution |
pvals |
p-values computed foreach correlation value using the Chi-square distribution |
Christian Montel
Becker, R. A., Chambers, J. M. and Wilks, A. R. (1988) The New SLanguage.Wadsworth & Brooks/Cole.
Hotelling H. (1936). Relations between two sets of variables. Biometrika, 28, p. 321–327.
Seber, G. A. F. (1984). Multivariate Observations. New York: Wiley, p. 506f.
Pedhazur E. (1997). Multiple regression in behavioral research Orlando: Harcourt Brace College Publishers, p. 939f.
pop <- LifeCycleSavings[,2:3] oec <- LifeCycleSavings[,-(2:3)] ccor <- cancor(pop,oec) ccorPvals(ccor,N=dim(LifeCycleSavings)[1],onlyPs=FALSE)