Date: Fri, 17 Mar 2000 11:51:16 -0500
From: "James Bailey" <James_Bailey@emory.org>
Subject: I am interested in using NONMEM in the analysis of binary data but I don't want to use the standard
I am interested in using NONMEM in the analysis of binary data but I don't want to use the standard logit or probit function. Instead I want to use a ramp model for the probability of drug effect, i.e.,
Prob=0 if Conc<C50-b
Prob=(Conc-(C50-b))/(2b) if C50-b<Conc<C50+b
Prob=1 if Conc>C50-b
where C50 and b are the parameters I wish to estimate.
NONMEM is not accepting the above definition of P. Any suggestions?
Jim Bailey
I am interested in using NONMEM in the analysis of binary data but I don't want to use the standard
Date: Fri, 17 Mar 2000 09:41:17 -0800 (PST)
From: ABoeckmann <alison@c255.ucsf.edu>
Subject: Re: I am interested in using NONMEM in the analysis of binary data but I don't want to use the standard
This should be doable with NONMEM. I suggest that you send your control stream and data file to the NONMEM consultants at Globomax, nmconsult@globomax.com. They should be able to help you fix whatever is wrong in your definition of P.
- Alison Boeckmann