assess model for logistic regression
Dear all, I am trying to run a PK/PD model with PD being binary data. After
looking into previous post I got my control stream running. My question is that
how to assess the model fit? Can I still use PRED, RES and WRES? Attached is
the code I found and used after a bit modification.
##########################################################
From: "Piotrovskij, Vladimir [JanBe]" <[email protected]>
Subject: RE: logistic regression
Date: Mon, 17 Sep 2001 14:05:46 +0200
Charlotte,
It is possible to solve some of your problems in NONMEM. However, the best
way is to apply generalized linear regression using one of the statistical
packages.
With NONMEM, try the following control stream:
$PROB dichotomous response: fixed effect of schedule
$DATA nmd.ssc
$INPUT ID AUC SCHD DV
; schedule coded as 1,2,3, etc.
$PRED
SCHD1 = 0
SCHD2 = 0
SCHD3 = 0
SCHD4 = 0
IF (SCHD.EQ.1) SCHD1 = 1
IF (SCHD.EQ.2) SCHD2 = 1
IF (SCHD.EQ.3) SCHD3 = 1
IF (SCHD.EQ.4) SCHD4 = 1
SLOPE = THETA(1)
E50 = SCHD1*THETA(2)+SCHD2*THETA(3)+SCHD3*THETA(4)+SCHD4*THETA(5)
INT = -LOG(E50) * SLOPE
LOGIT = INT + SLOPE * LOG(AUC) + ETA(1)
A=EXP(LOGIT)
P=A/(1+A)
IF (DV.EQ.1) Y=P
IF (DV.EQ.0) Y=1-P
$THETA
(2 5 7); 1 SIGM
(0 30 50); 2 E50 SCHD=1
(20 50 70); 3 E50 SCHD=2
(40 70 90); 4 E50 SCHD=3 (60 100 200); 5 E50 SCHD=4
$OMEGA .0001
$EST METHOD=COND LAPLACE LIKE MAX=500 PRINT=10
$COV
$TABLE ID AUC SCHD DV FILE=tab.ssc ONEHEADER NOPRINT