principal components - correlation matrix?

[chartwel]

Hi, I'm using eviews 7 and am using a principal components analysis on time series, coded data. I have successfully extracted PCs using eviews, and have narrowed down the components that I want to use based on their eigenvalues. However, I want to get a correlation matrix of correlations between these new components and the original variables that were included - how can I do that in eviews? Basically, i have 21 variables of central bank independence that I've extracted 7 principal components from, and I want to see the correlation matrix between those 7 PCs and the original 21 variables. How can I do that?

[EViews Glenn]

Create a group object with all 28 variables and compute the correlation for that group of series.

[chartwel]

ok, how do I get those 7 principle components in a usable form to be included in a group object though? Basically, I use the command after opening the 21 variables as a group to "view/principal components," input my specifications, and then it spits out the list of PCs. How do I get that output saved so that I can use it in a group object?

Cheers,
Chris

[EViews Glenn]

I had thought you had already made the PC scores. Go to Proc/Make Principal Components... in the original group menu.