ProgId Word.Document Generator Microsoft Word 10 Originator Microsoft Word
10 Dear NONMEM users,
I would like to investigate the individual contributions to the objective
function value. I have seen the ROCM50 example in the HELP guide (using
$PRED).
To be applicable to user supplied $PK or $DES (e.g. ADVAN4 or ADVAN6 routines)
as well, I guess I have to use a $INFN record?
I have tried the following with success (at least I get some estimates of the
individual OFV contributions. .)
$PK
V=THETA(1)*EXP(ETA(1))
CL=THETA(2)*EXP(ETA(2))
S1=V
$ERROR
IPRED=F
W=(F*F*THETA(3)**2+THETA(4)**2)**0.5
Y=F+EPS(1)*W
IRES=DV-IPRED
IWRES=IRES/W
$INFN
" FIRST
" COMMON /ROCM50/ IIDX(1000),CNTID(1000)
" DOUBLE PRECISION CNTID
" INTEGER IIDX
IF (ICALL.EQ.3) THEN
" DO 100 I=1,NINDR
" WRITE (70,*) IIDX(I),CNTID(I)
" 100 CONTINUE
ENDIF
Could someone please tell me whether this is correct or not?
Regards
Daniel Röshammar
Unit for Pharmacokinetics and Drug Metabolism
Sahlgrenska Academy at Göteborg University
Phone: +46 31 773 34 94
individual contributions to OFV
It looks correct. Add up the numbers in fort.70 - the sum of the
individual contributions should equal
the NONMEM objective function as displayed in the output report.
On Tue, 13 Feb 2007 16:38:45 +0100, ""Daniel Röshammar""
<[EMAIL PROTECTED]> said:
> ProgId Word.Document Generator Microsoft Word 10 Originator Microsoft
> Word 10 Dear NONMEM users,
>
> I would like to investigate the individual contributions to the objective
> function value. I have seen the ROCM50 example in the HELP guide (using
> $PRED).
> To be applicable to user supplied $PK or $DES (e.g. ADVAN4 or ADVAN6
> routines) as well, I guess I have to use a $INFN record?
>
> I have tried the following with success (at least I get some estimates of
> the individual OFV contributions. .)
>
> $PK
> V=THETA(1)*EXP(ETA(1))
> CL=THETA(2)*EXP(ETA(2))
> S1=V
> $ERROR
> IPRED=F
> W=(F*F*THETA(3)**2+THETA(4)**2)**0.5
> Y=F+EPS(1)*W
> IRES=DV-IPRED
> IWRES=IRES/W
> $INFN
> " FIRST
> " COMMON /ROCM50/ IIDX(1000),CNTID(1000)
> " DOUBLE PRECISION CNTID
> " INTEGER IIDX
> IF (ICALL.EQ.3) THEN
> " DO 100 I=1,NINDR
> " WRITE (70,*) IIDX(I),CNTID(I)
> " 100 CONTINUE
> ENDIF
>
> Could someone please tell me whether this is correct or not?
>
> Regards
> Daniel Röshammar
> Unit for Pharmacokinetics and Drug Metabolism
> Sahlgrenska Academy at Göteborg University
> Phone: +46 31 773 34 94
--
Alison Boeckmann
[EMAIL PROTECTED]