[R] Correlation matrices

[Charles C. Berry]

"Patrik Waldmann" <Patrik.Waldmann at sysbot.lu.se> asked:

PW> are there any good methods in R that will test if two correlation
PW> matrices (obtained in different ways) are equal? Something better than
PW> the Mantel test would be preferable.

and from the reponses, it would appear that the answer is 'no'.

However, the methods outlined in:

Beran, R. and M Srivistava (1985) Bootstrap tests and confidence 
regions for functions of a covariance matrix, Annals of Statistics 
13:95-115, 1985.

would be easy enough to implement and do not require MVN
assumptions. 

All that is required is to estimate the covariance structure under the
desired null model, to compute a statistic that measures departures
from the null, and the computation of matrix square roots, if I
recall.

-- 
Charles C. Berry                        (858) 534-2098 
                                         Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	         UC San Diego
http://hacuna.ucsd.edu/members/ccb.html  La Jolla, San Diego 92093-0645

 
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._