NONMEM Users Network Archive

Hosted by Cognigen

RE: simulation question

From: Elodie Plan <elodie.plan>
Date: Fri, 6 Feb 2009 11:52:08 +0100

Dear Ethan,


Perl-speaks-NONMEM (PsN) is a free program developed by Uppsala Research
Pharmacometrics group. We recently added a module performing Stochastic
Simulations followed by re-Estimations (SSE). The purpose is to investigate
aspects related to simulation or estimation methods, models, study designs
or statistical inference. I think it will solve your problem.

You can download it here as well as find the sse userguide:

What it does compared to NSUB=X is to run all problems in parallel, so a
termination won't affect the others. And moreover it provides a summary with
computations like bias etc... A simple single command will do the simulation
and the re-estimation with the different designs: sse run1.mod -samples=200

You can contact me or my colleagues for more details.


Best regards,



            Elodie L. Plan, PharmD, MSc,



Div. of Pharmacokinetics and Drug Therapy,

 Department of Pharmaceutical Biosciences,
  Faculty of Pharmacy, Uppsala University

      Box 591 - 751 24 Uppsala - SWEDEN

Office +46 18 4714385 - Fax +46 18 4714003


From: owner-nmusers
Behalf Of Ethan Wu
Sent: Thursday, February 05, 2009 11:19 PM
To: nmusers
Subject: [NMusers] simulation question


Dear users,

  I am trying to compare several specific PK/PD study designs by:

 -- run 200 simulations with the final model (develope from original

 -- fit the final model to the 200 simulated dataset

To achieve above, I used $SIM SUBPROB=200 option

 however, nonmem would completely stop after running into estimation
problem at one specific simulation/estimation cycles, for some designs it
stop even before 10th iterations.

Is there anyway nonmem could continue go on?

Or, does someone know alternative way to achieve the goal?




Received on Fri Feb 06 2009 - 05:52:08 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: