Bookmark and Share

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: Testparm Command in Logistic Regression Analysis


From   Alfonso Sanchez-Penalver <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: Testparm Command in Logistic Regression Analysis
Date   Wed, 27 Nov 2013 08:23:16 -0500

Hola Isabel,

I am curious as to why you have different number of observations in the restricted (nested) model rather than in the unrestricted model. All you're doing is restricting coefficients to be zero, and this doesn't necessarily imply dropping observations. The -testparm- test is based on the unrestricted model and its observations. Could you please clarify this?

Thanks,

Alfonso Sanchez-Penalver

> On Nov 27, 2013, at 7:54 AM, <[email protected]> wrote:
> 
> Thanks Marteen, this seems to work!
> Just in general, is it correct to use the test/testparm command in logistic regression or does it only apply for linear regression models?
> As I already stated I can't use the lrtest command to compare nested models because I have  different numbers of observations within the models. So what's the appropriate way to test the overall significance of a variable in a logistic regression model (and then to decide whether it stays in the model or not)?
> 
> Thank you for your help!
> 
> Isabel
> 
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of Maarten Buis
> Sent: Mittwoch, 27. November 2013 13:40
> To: [email protected]
> Subject: Re: st: Testparm Command in Logistic Regression Analysis
> 
>> On Wed, Nov 27, 2013 at 11:16 AM,  <[email protected]> wrote:
>> I'm performing a logistic regression with 5 categorical predictor variables.
>> 
>> xi: logit status2 i.season2 i.sex2 i.age2 i.unit2 i.distance5
>> 
>> I intended use the- testparm- command  for each variable to get an overall significance for the variables in the model.
>> 
>> testparm i.season2
>> testparm i.sex2
>> 
>> Both of these commands are not working, as Stata tells me "no such variables".
> 
> You should not use the -xi:- prefix anymore. It is only there for some
> (user-written) programs that do support the factor variable notation (yet), and for backward compatability for older .do-files. All you need to do to fix your code is leave out the -xi:- prefix.
> 
> logit status2 i.season2 i.sex2 i.age2 i.unit2 i.distance5 testparm i.season2 testparm i.sex2
> 
> Hope this helps,
> Maarten
> 
> 
> ---------------------------------
> Maarten L. Buis
> WZB
> Reichpietschufer 50
> 10785 Berlin
> Germany
> 
> http://www.maartenbuis.nl
> ---------------------------------
> *
> *   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/

*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index