LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (February 2008, week 2)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:   Tue, 12 Feb 2008 13:52:34 -0500
Reply-To:   IKF <ikf@HOTMAIL.COM>
Sender:   "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:   IKF <ikf@HOTMAIL.COM>
Subject:   The Same NLMIXED question

I also have the stupid question of figuring out whether I need to consider this binary or binomial. The top is an example that I was looking at from the archives, while the bottom is what I have been working with. My outcome (culture) can only be 0 or 1. I think that p gives me the log(OR) and estimate gives me the OR. Am I even on the right track??? Thanks!

proc nlmixed data=test2; eta = beta0 + gamma0; p = exp(eta)/(1+exp(eta)); prob = comb(n,y) * (p**y) * (1-p)**(n-y); loglike = log(prob); model y ~ general(loglike); random gamma0 ~ normal([0], [vgamma0]) subject=cluster; run;

proc nlmixed data=pneumo1; parms b0=-1 b1=0 b2=0 b3=0 b4=0 s2=0 ; teta = b0 + b1*babyagedays + b2*babyagedays1 + b3*babyagedays2 + b4*babyagedays3 + u; p = exp(teta) / (1 + exp(teta)); model culture ~ binary(p); random u ~ normal(0,s2) subject=pin; estimate 'exp(b1)' exp(b1); run;


Back to: Top of message | Previous page | Main SAS-L page