NONMEM Users Network Archive

Hosted by Cognigen

Re: How to scale PARAMETERs

From: Nick Holford <n.holford>
Date: Wed, 27 Aug 2008 08:54:54 +1200

Hi,

The almost incomprehensible display of parameter estimates in their
internally scaled format has been a long standing 'feature' of NONMEM
and one that has been suggested for change many times. Perhaps in NONMEM 7?
The 'solution' which I think appeared in NONMEM V (described below by
Bill) is a quick and dirty fix which is rather clumsy for the user to use.

Wings for NONMEM users can get access to these intermediate results and
update the control stream used to start the run with the latest values
with the command:
    nmctl /runname/ i
The second argument 'i' tells nmctl to get values from a file named
INTER.txt in the WFN run results folder. This is not the default name
for the file but its name can be easily changed in BLKDAT.for i.e.

      DATA FNSTP,FNINT,FNERR,FNINTR,FNULL
     1/'nul','con','PRDERR','INTER.txt','nul'/

This change makes it easier for Windows users to use their text editors
to read the file because it has a meaningful extension.

You can stop the INTER file being deleted at the end of the run by this
change in CFILES.for:

C Stop INTER being deleted at end of run.
      CLOSE (UN(28))

These changes are made automatically if you use the NMQUAL based NMQ4WFN
package to install NONMEM for use with WFN.

Best wishes,

Nick

Bachman, William wrote:
> You can find the unscaled parameters at the iterations requested in
> the PRINT= option of $EST in the INTER file if you specify MSFO as an
> $EST option. The file exists only during the run (by default).
>
> ------------------------------------------------------------------------
> *From:* owner-nmusers
> [mailto:owner-nmusers
> *axel.facius
> *Sent:* Tuesday, August 26, 2008 12:45 PM
> *To:* nmusers
> *Subject:* [NMusers] How to scale PARAMETERs
>
> Hi there!
>
> During $ESTimation, NONMEM prints PARAMETER values. At iteration 0,
> these are all 'scaled' to 0.1. During long NONMEM runs I'm sometimes
> interested in seeing where my current estimate of parameters are, so I
> would like to re-scale them back. Unfortunately simple re-scalings
> like "parameter" * 10 * "Initial-estimate" do not work (except for
> iteration 0 ;-). If I do this for the parameters in the last
> iteration, I do not end up with the final parameter estimates..?!
>
> Does anybody know how to re-scale?
>
> Cheers, Axel
>
> Here some example:
>
> initial est.: 0.4715E+00
> initial par.: 0.1000E+00
> final par.: -0.5854E+00
> final est.: 2.38E-01
>
> taken from:
>
> [...]
>
> 0INITIAL ESTIMATE OF THETA:
> LOWER BOUND INITIAL EST UPPER BOUND
> 0.0000E+00 0.4715E+00 0.1000E+07
> [...]
>
> MONITORING OF SEARCH:
>
> 0ITERATION NO.: 0 OBJECTIVE VALUE: 0.65915E+05 NO. OF FUNC.
> EVALS.:20
> CUMULATIVE NO. OF FUNC. EVALS.: 20
> PARAMETER: 0.1000E+00 ..
> GRADIENT: 0.1145E+06 ...
>
> [...]
>
> 0ITERATION NO.: 61 OBJECTIVE VALUE: -0.39051E+04 NO. OF FUNC.
> EVALS.:36
> CUMULATIVE NO. OF FUNC. EVALS.: 2195
> PARAMETER: -0.5854E+00 ...
> GRADIENT: 0.3601E+02 ...
> 0MINIMIZATION SUCCESSFUL
>
> [...]
>
> ************************************************************************************************************************
> ********************
> ********************
> ******************** FINAL PARAMETER
> ESTIMATE ********************
> ********************
> ********************
> ************************************************************************************************************************
>
>
>
> THETA - VECTOR OF FIXED EFFECTS PARAMETERS *********
>
>
> TH 1 ...
>
> 2.38E-01 ...
>
>
> ----------------------------------------------------------------------
> Proprietary or confidential information belonging to Nycomed Group may
> be contained in this message. If you are not the addressee indicated
> in this message, please do not copy or deliver this message to anyone.
> In such case, please destroy this message and notify the sender by
> reply e-mail. Please advise the sender immediately if you or your
> employer do not consent to Internet e-mail for messages of this kind.
> Opinions, conclusions and other information in this message that
> pertain to the sender's employer and its products and services
> represent the opinion of the sender and do not necessarily represent
> or reflect the views and opinions of the employer.
> ----------------------------------------------------------------------
>
> ICON plc made the following annotations.
> ------------------------------------------------------------------------------
> This e-mail transmission may contain confidential or legally privileged information
> that is intended only for the individual or entity named in the e-mail address. If you
> are not the intended recipient, you are hereby notified that any disclosure, copying,
> distribution, or reliance upon the contents of this e-mail is strictly prohibited. If
> you have received this e-mail transmission in error, please reply to the sender, so that
> ICON plc can arrange for proper delivery, and then please delete the message.
> Thank You,
> ICON plc
> South County Business Park
> Leopardstown
> Dublin 18
> Ireland
> Registered number: 145835
>
> ***Please note that starting Monday July 14th ICON has standardized on the iconplc.com domain name***
>
>
>

--
Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand
n.holford
http://www.fmhs.auckland.ac.nz/sms/pharmacology/holford

Received on Tue Aug 26 2008 - 16:54:54 EDT

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to: nmusers-request@iconplc.com.

Once subscribed, you may contribute to the discussion by emailing: nmusers@globomaxnm.com.