[NMusers] RE: re: bug in NONMEM VI?

 

________________________________

From: John Lukas
Sent: Mon 12/17/2007 11:51 PM
To: nmusers_at_globomaxnm.com
Subject: re: bug in NONMEM VI?



Dear Nmusers,

 

A simple LME model run using NM version VI, FO method, with 4 covariates =
(5 THETA's) as,

.

.

$PRED

 

SUM1=THETA(2)*COV1+THETA(3)*COV2

SUM2=THETA(4)*COV4+THETA(5)*COV3

 

TLAU=THETA(1)+SUM1+SUM2

F=TLAU+ETA(1)

 

Y=F+EPS(1)

 

$THETA 4 .1 .1 .1 .1

$OMEGA .1

$SIGMA .1

 

$ESTIMATION MAXEVALS=1900 PRINT=5 METHOD=0

 

(let's call this FIT A) gave a different fit from another (FIT B) where =
only the order of THETA( 4) and 5 was reversed, all else kept the same, =
as,

.

.

SUM2=THETA(5)*COV4+THETA(4)*COV3

.

.

(even more strange, FIT B, the reversed order run, proved to be the good =
run!)

 

NONMEM version V had the same result for both FITS A and B, as expected =
(and same as FIT B from version 6). FOCE gave good answers always in =
both NONMEM version V and version VI. But, the question remains about =
that dependence on ordering of the THETAs with the FO method for version =
VI.

 

Any comments? Has there been discussion on this earlier that I missed? =
Thanks in advance.

 

John

PS I am aware of other *important* incosistencies between version VI and =
V from university colleagues working on a PK problem with FOCE...

 

 

John C Lukas

Strategic Consulting Services

Pharsight Corp.

 

line: + 33 492 726 495

cell: + 33 626 496 777

 

 


Received on Wed Jan 02 2008 - 14:35:46 EST