Author: Moshe Adrian <Moshe.Adrian@morganstanley.com>
Date: Fri, 11 Jul 2003 13:32:58 -0400
Dear S-plus experts, I am having some trouble with Factor analysis in my version of S-Plus, S-Plus 2000 for Windows NT (although the same problem should appear in later versions). When I perform fact
Author: Moshe Adrian <Moshe.Adrian@morganstanley.com>
Date: Fri, 11 Jul 2003 12:51:27 -0400
Dear S-plus experts, I am having some trouble with Factor analysis in my version of S-Plus, S-Plus 2000 for Windows (although the same problem should appear in later versions). When I perform factor
Author: Moshe Adrian <Moshe.Adrian@morganstanley.com>
Date: Fri, 11 Jul 2003 16:37:59 -0400
I used factanal, yes. I just tried the "mle" method and the negative uniqueness disappears. I am going to investigate the mle method, and see if this is ok to use for my purposes. Thanks Spencer, Mos
Author: Moshe Adrian <Moshe.Adrian@morganstanley.com>
Date: Mon, 14 Jul 2003 16:05:47 -0400
Thanks Spencer, Is there a likelihood ratio test function in S-Plus and how can it be interpreted? (I am now using method = "mle" for my factor analysis) Moshe
Author: Moshe Adrian <Moshe.Adrian@morganstanley.com>
Date: Tue, 15 Jul 2003 13:19:25 -0400
Thank you. The only thing left for me to do is to compare this with S-plus's How do you perform a test of the hypothesis that the specified number of factors is adequate to explain the model directly