Barber,
Two options to consider:
1- You can compare both models over observations with no missing values on
FUMATIPO.
Just modify your if command in the old model. Simple way but loss of 242
observations
2- You can add a variable (MISSING(FUMATIPO)) in interaction with your old
and new model
Old model = OldModel1 if (missing FUMATIPO)) OldModel2 (if not missing
FUMATIPO) whatever FUMATIPO value is
New model = NewModel1 if (missing FUMATIPO)) NewModel2 (if missing FUMATIPO
+ i.FIMATIPO)
More complexe, possible overparam. but you'll have all the observations
Best
Naji
-----Message d'origine-----
De : [email protected]
[mailto:[email protected]]De la part de Xavi Barber
Envoye : lundi 24 novembre 2003 13:50
A : [email protected]
Objet : st: Missing Value & Logistic regression
Dears stata users:
I'm doing a logistic regression, and now i have ajusted by sex,
age and others variables.
My question is:
My old model: xi:logistic esofago i.sexo1 i.edad3c
i.estudi3c i.tipobeb2 i.fuma1 if fil_bebe==1
My new model: xi:logistic esofago i.sexo1 i.edad3c
i.estudi3c i.tipobeb2 i.fuma1 i.fumatipo if fil_bebe==1
but the problem is that FUMATIPO have 242 missing values.
I need to put the result in a typical journal table -->
Case - Control ; Mean (stad. desv) ; OR IC(OR)
but, i can to compare the 1st model vs then 2nd model??
thanks
*
* 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/
