topic Re: GENMOD dropping variable in parameter estimates table in Statistical Procedures

GENMOD dropping variable in parameter estimates table

lsandell — Tue, 10 May 2022 16:01:24 GMT

Hi there, I am modeling an outcome with multiple covariates (GENMOD, logistic regression). However, my parameter estimates table is dropping out an entire variable from the table that is in my model. I know there are no missing observations because I have selected only complete cases in my dataset.

Hoping to understand why I'm not getting the variable 'fever' in my final output estimates table? Also, if it's not in the table, is it at least being adjusted for in the results?

Thanks.

proc genmod data = hospital;
	class	age (ref='X')
			pathogen_new (ref='X')
			diarrhea (ref='NO') 
			bloodydiarr (ref='NO') 
			fever (ref='NO') 
			specimen (ref='X')
			 / param=ref;
	model 	hospital (event='YES') = age  pathogen_new diarrhea bloodydiarr fever / dist = bin link = log;
	format age Agefull. specimen $source_cllp.;
run;

Re: GENMOD dropping variable in parameter estimates table

Reeza — Tue, 10 May 2022 16:15:49 GMT

Show the log.

Re: GENMOD dropping variable in parameter estimates table

SteveDenham — Mon, 16 May 2022 15:08:16 GMT

The output reflects the model. You have

age  pathogen_new diarrhea bloodydiarr fever

as your right hand side variables, and that is what shows in your output. Until you add "specimen" to the model, it will not appear in the output.

I suspect that fever is confounded completely with at least one other variable (this looks like a good question for @StatDave ).

SteveDenham