2
$\begingroup$

Is it possible to predict values of variables in a matrix Y from values of variables in a matrix X, ie. using canonical correlations since some variables are correlated among these two matrix?

X and Y contain variables for the same individuals.

If possible, I would like an example using R.

Edit: the aim is to fill missing values in the matrix Y from correlated variables that are in matrix X.

$\endgroup$
  • $\begingroup$ Is this related to one of your earlier question? $\endgroup$ – chl Sep 8 '11 at 14:03
  • $\begingroup$ Are you sure you want to use canonical correlation? Where does your data come from? Maybe canonical correlation isn't the best way to proceed: the aim is to fill missing values in the matrix Y from correlated variables that are in matrix X $\endgroup$ – deps_stats Sep 8 '11 at 17:39
  • $\begingroup$ Maybe another technique will be more adequate. My data are datas collected from user using a service that I provide. I want to use data collected from users (values of variables in matrix X) to fill missing values about them for the variables that are in matrix Y. $\endgroup$ – hijack Sep 9 '11 at 9:34
  • $\begingroup$ @chl. absolutely $\endgroup$ – hijack Sep 9 '11 at 10:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.