Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Factor analysis
From
William Buchanan <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: Factor analysis
Date
Fri, 13 Sep 2013 06:34:57 -0500
This doesn't seem to be a "factor analysis" issue per se. You can observe the diagnoses, so why not just create an indicator using the observed diagnoses directly (e.g., if the ith observation has >1 diagnosis you label them a comorbid patient)? If you need to use some type of latent variable modeling, it would probably make more sense to use some form of a latent class model and use the predicted class memberships to define observations rather than some single continuous value that isn't directly related to classifying observations.
HTH,
Billy
Sent from my iPhone
On Sep 13, 2013, at 6:12, [email protected] wrote:
> Dear statlisters
> I have a problem and i need help from you.
> I´ve made a fctor analysis of a hugue database (300000 obs) of medical diagnosis in order to unveil teg relationship betrween the diagnostics.
> I´ve found 3 factors that explain well the multimorbidity of the patients.
> Now i need to create a variable (0/1) that saus iof a person has multimorbidity or not. The condition to be multimorbidity is to have loadigs of .30 or more in at least three of the diagnositcs, in any one of the factors.
> So i have about 60 variables with diferent loadings and i need a variable in the previous conditions.
> Does anyone can help.
> Thanks in advance
> Antonio Sousa
> Portugal
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/