NONMEM Users Network Archive

Hosted by Cognigen

RE: [NMusers] model of error, linearisation, additive error, and VPC

From: Huub Jan Kleijn <>
Date: Sun, 31 Jan 2016 18:02:08 +0000

Hi Pavel,

I guess I would opt for using logtransformed concentrations as dv in your m=
odel. That would deal with the negative model predictions.

From: [] on behalf=
 of Pavel Belo []
Sent: Friday, January 29, 2016 10:47 PM
Cc: ''
Subject: [NMusers] model of error, linearisation, additive error, and VPC

Hello NONMEM Users,

When I tried to print log-scaled VPC, there was an error message about nega=
tive values. It can be caused by an additive error and/or linearization of=
 the error model when NONMEM transforms Y = F*DEXP(ERR(1)*SD1) into Y =
 = F+ F*ERR(1)*SD1. The error is inflated at small concentrations and re=
moving the additive term is not an option in my case.

Is there an easy way to solve it? It can be something like Y = F**GAMMA=
*DEXP(ERR(1)*SD1) or
Y = F+ F*ERR(1)*SD1 + F**GAMMA*ERR(2)*SD2, where GAMMA<1.

Received on Sun Jan 31 2016 - 13:02:08 EST

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to: Once subscribed, you may contribute to the discussion by emailing: