Namespaces
Variants
Actions

Multiple-correlation coefficient

From Encyclopedia of Mathematics
Revision as of 17:01, 7 February 2011 by 127.0.0.1 (talk) (Importing text file)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

A measure of the linear dependence between one random variable and a certain collection of random variables. More precisely, if is a random vector with values in , then the multiple-correlation coefficient between and is defined as the usual correlation coefficient between and its best linear approximation relative to , i.e. as its regression relative to . The multiple-correlation coefficient has the property that if and if

is the regression of relative to , then among all linear combinations of the variable has largest correlation with . In this sense the multiple-correlation coefficient is a special case of the canonical correlation coefficient (cf. Canonical correlation coefficients). For the multiple-correlation coefficient is the absolute value of the usual correlation coefficient between and . The multiple-correlation coefficient between and is denoted by and is expressed in terms of the entries of the correlation matrix , , by

where is the determinant of and is the cofactor of ; here . If , then, with probability , is equal to a linear combination of , that is, the joint distribution of is concentrated on a hyperplane in . On the other hand, if and only if , that is, if is not correlated with any of . To calculate the multiple-correlation coefficient one can use the formula

where is the variance of and

is the variance of with respect to the regression.

The sample analogue of the multiple-correlation coefficient is

where and are estimators of and based on a sample of size . To test the hypothesis of no relationship, the sampling distribution of is used. Given that the sample is taken from a multivariate normal distribution, the variable has the beta-distribution with parameters if ; if , then the distribution of is known, but is somewhat complicated.

References

[1] H. Cramér, "Mathematical methods of statistics" , Princeton Univ. Press (1946)
[2] M.G. Kendall, A. Stuart, "The advanced theory of statistics" , 2. Inference and relationship , Griffin (1979)


Comments

For the distribution of if see [a2], Chapt. 10.

References

[a1] T.W. Anderson, "An introduction to multivariate statistical analysis" , Wiley (1958)
[a2] M.L. Eaton, "Multivariate statistics: A vector space approach" , Wiley (1983)
[a3] R.J. Muirhead, "Aspects of multivariate statistical theory" , Wiley (1982)
How to Cite This Entry:
Multiple-correlation coefficient. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Multiple-correlation_coefficient&oldid=12840
This article was adapted from an original article by A.V. Prokhorov (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article