Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: Negative eigen values in factor, pf command?


From   "Martin Weiss" <martin.weiss1@gmx.de>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: RE: Negative eigen values in factor, pf command?
Date   Wed, 29 Apr 2009 00:21:14 +0200

<>

"PS. if "Thanks emails" are inappropriate for the listserv,
please let me know off list and I'll refrain in the future...as I feel
I will be asking more questions soon)"


We can tell you this on the list, as it is plain to see from the FAQ http://www.stata.com/support/faqs/res/statalist.html#others
toward the end:

"Continuing or closing a thread you started is important, especially by answering secondary questions and by reporting what solved your problem. You can then thank those who tried to help. "

I hope that you will continue asking questions on the list :-)


HTH
Martin
_______________________
----- Original Message ----- From: "Jean-Gael Collomb" <JG@ufl.edu>
To: <statalist@hsphsun2.harvard.edu>
Sent: Wednesday, April 29, 2009 12:15 AM
Subject: Re: st: RE: Negative eigen values in factor, pf command?


Thanks to all for the multiple and prompt answers. The latter posts are losing me, but they may be of interest to more statistically curious researchers (by the way, I did also try a factor, ml command, and the results also seemed different than the pca results). The basic explanation of what was going on with the factor command was helpful to understand the negative eigen values and unblocked the discussion with my statistician colleagues. I basically wanted to check the dimensionality of an index for a latent concept. I am going to put more stock in the pca results, and after looking at the factor loadings, I feel in a better position to interpret the results. Thanks (PS. if "Thanks emails" are inappropriate for the listserv, please let me know off list and I'll refrain in the future...as I feel I will be asking more questions soon)
Cheers,

Jean-Gael "JG" Collomb
PhD candidate
School of Natural Resources and Environment / School of Forest Resources and Conservation
University of Florida
jgcollomb@gmail.com
jg@ufl.edu
+1 (352) 870 6696




On Apr 28, 2009, at 3:43 PM, Verkuilen, Jay wrote:

Michael I. Lichter wrote:

So what are you advocating instead? ML, which isn't very robust to
weirdness (technical term) in the data? Or something other than >FA/ PCA?

No, I'm following the research of people like Rod McDonald (retired to
his beloved Sydney), Albert Maydeu-Olivares (Barcelona) and Bud  McCallum
(UNC-Chapel Hill), who support the use of the Unweighted Least Squares
criterion function, or multi-stage estimation methods that uses ULS
followed by a few iterations of ML or GLS. ULS is mathematically
equivalent to MINRES. Stata doesn't have ULS/MINRES (alas).

One of the downsides to ULS is the fact that it doesn't generate
standard errors and GoF statistics, but Albert's been rectifying that
situation in a sequence of articles in various journals  (Psychometrika,
JASA, etc.): http://www.ub.edu/gdne/psicometriaen.html.



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index