st: RE: Factor Analysis of Individually Matched Case Control Study with collinearity problem

["Kieran McCaul" <Kieran.McCaul@uwa.edu.au>]

...

have you tried -exlogistic-?  

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of alexandra Clarke
Sent: Wednesday, 4 August 2010 5:04 PM
To: statalist@hsphsun2.harvard.edu
Subject: st: Factor Analysis of Individually Matched Case Control Study with collinearity problem

Dear All,
I was wondering how to go about doing a Factor Analysis of an individually matched Case-Control study in STATA 11 with one control per case and 25 cases. I am unable to build a multivariable model using Conditional Logistic Regression with more than 3 terms due I think to a collinearity problem and so have not been able to assess for confounding and interaction properly. I have performed a factor analysis on the whole data set, not taking into account the matching, which has helped me to merge some variables. My outcome is binary, obviously, but I have binary, categorical, ordinal and continuous explanatory variables. Also, what is the best way of assessing collinearity with this type of data. I made a table of all my explanatory variables relationships with each other using Conditional Logistic Regression, which has again helped partially, and have tried the VIF and COLLIN commands, but don't really know what I am doing and need to account for matching. 
With best wishes,
A

 




      

*
*   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/