Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: factor analysis - interpreting the results, inclusion of dp variable


From   "georg wernicke" <georg.wernicke@googlemail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: factor analysis - interpreting the results, inclusion of dp variable
Date   Mon, 13 Aug 2007 18:44:35 +0100

Dear SR Mills,

thank you very much for your help. It is most appreciated. I am just
going over the mentioned article, it seems as i have some
collinearity. i dropped 2 variables which are indicated as collinear
by the belsley threshold, the factortest immideately went up from 4 to
7 factors and will prob. improve once i have corrected for all
possibly collinear variables.

however, do i include the dependent variable in the factor equation?
(e.g. factor dependent independent1 independent2 etc..)

and what do i get from knowing that the factor analysis gives me lets
say 9 factors?

once again, thanxs for your help

george

On 8/13/07, SR Millis <srmillis@yahoo.com> wrote:
> Based on guidelines from Belsley et al. (pp.
> 112-113;2004) it does appear to me that you do have
> significant collinearity among some of your variables.
> Large variance-decomposition proportions (>.5)
> associated with each condition index (>30) identifies
> the offending variables.  I'd want to drop some of the
> variables causing the collinearity---and re-run the
> analysis--with the individual variables.
>
> SR Millis
>
> Belsley, D. A., Kuh, E., & Welsch, R. E. (2004).
> Regression diagnostics: Indentifying influential data
> and sources of collinearity. Hoboken, NJ: John Wiley &
> Sons.
>
> --- georg wernicke <georg.wernicke@googlemail.com>
> wrote:
>
> > Hello SR Millis,
> >
> > the highest condition index is 57.29. regarding the
> > variance
> > decomposition proportions, there are 6 above 0.5
> > when the condition
> > index is ~30. . seems a bit weird that the table
> > using coldiag2 gives
> > me 17 indexes even so i only have 16 variables.
> >
> > thanks again,
> >
> > george
> >
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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