Pavel:
The objective function progress looks good. You should expect some
Monte Carlo fluctuations. You should also run more iterations (perhaps
NITER=200), and set CTYPE=3, which turns on the termination tester. To
resume where you left off, rename your new control stream file, and put
in the
-0769
Email: robert.ba...@iconplc.com
Web: www.icondevsolutions.com
From: non...@optonline.net [mailto:non...@optonline.net]
Sent: Friday, October 30, 2009 11:03 AM
To: Bauer, Robert
Cc: nmusers@globomaxnm.com
Subject: Re: RE: RE: [NMusers] method=ITS
If you get this error only occasionally, and the problem still runs, it
may not be in issue. Alternatively, you can try different TOL settings.
Finally, if these do not work, you should fall back on advan9.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics
ICON Development Solutions
: Sam Liao [mailto:sl...@pharmaxresearch.com]
Sent: Wednesday, November 04, 2009 9:04 PM
To: Bauer, Robert; 'nmusers'
Subject: RE: [NMusers] NM7 question
Dear Robert:
Thanks for your kindly reply. I tried either one of the following two
methods to run M3 method to handle BQL samples, it failed
Try
$COVARIANCE UNCONDITIONAL
This will give you a standard error from the METHOD=IMP EONLY=1 step
even if there is some positive indefiniteness. But, be cautious about
its quality when such a message occurs.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics
ICON Development
I would recommend the following:
1. Set PRINT=
This way, iteration 0 is printed, as well as the final results.
Unlikely you will have more than iterations to have any intervening
iterations printed.
2. Do not specify the FILE= in a multi-problem statement such as the
$SIML NSUB=3
Ethan:
We have noticed this occurs with IMP on a few other problems. We have
modified the code so this occurs less often, and the change will be in
NONMEM 7.2, scheduled for January. Meanwhile, as Sebastian suggested,
only SAEM needs an IMP EONLY=1 step. The importance sampler itself
produces
Bill:
I investigated the problem and found that not verbatimizing the code
allows NM7/NMTRAN to properly construct the PRINT statement and set up
the connection of ID to EVTREC():
PRINT*,ID
When verbatimizing:
PRINT*,ID
NM7/NMTRAN does not connect ID with EVTREC()/DATREC(),
Dieter:
You can trust the fit. The negative eignvalue diagnostic arises from
evaluating the information matrix of the estimates evaluated after the fit.
Because this was constructed with Monte Carlo components, on occasion the
slight imprecision from calculating obscure off-diagonal elements
: Fidler,Matt,FORT WORTH,RD [mailto:matt.fid...@alconlabs.com]
Sent: Tuesday, September 14, 2010 5:14 PM
To: Bauer, Robert
Cc: nmusers@globomaxnm.com
Subject: RE: [NMusers] is there a way to change fit to csv
Bob,
I was looking further into the FORMAT statement. In the NONMEM 7 help
From: Ethan Wu [mailto:ethan.w...@yahoo.com]
Sent: Thursday, September 16, 2010 12:19 PM
To: Bauer, Robert; nmusers@globomaxnm.com
Subject: Re: [NMusers] quality of SE estimate by IMP method
Hi Bob,
Is there a way to pick up from the last run with SAEM/IMP so to only
run the additional step
From: Denney, William S. [mailto:william_den...@merck.com]
Sent: Tuesday, September 21, 2010 10:02 AM
To: Bauer, Robert
Subject: NM7.2 Parallelization and Virtualization
Hi Bob,
I hope that you're doing well these days. I have a couple of questions
Nick:
It means that a positive definiteness correction on the eta Hessian
matrix of a particular individual at a particular iteration was not
successful, but NONMEM goes ahead anyway. If it happened on an
intermediate iteration, and the error signal was not continually
repeated, then NONMEM
Dear Dr. Frede:
I too obtain an error from the Laplace method when correcting example10.ctl
with a FIX option on the prior to $OMEGA. This does not reflect an
installation issue, only an issue with the Laplace method, in that it
occasionally is very sensitive to starting positions. In our
Pavel:
If you do not mind sharing your control stream file, and if necessary,
your data file (or a truncated version of it), perhaps you could send it
to me. I may be able to determine if there is a way to modify the
control stream code to avoid these errors, or if there is a bug.
Robert J.
MCMC automatically sets ISAMPLE=1 for its run. ISAMPLE1 is only for
SAEM. To perform multiple chains, you must run separate control stream
problems with different initial SEED settings.
-Original Message-
From: owner-nmus...@globomaxnm.com [mailto:owner-nmus...@globomaxnm.com]
On Behalf
Dear nmusers:
This past week we have begun shipping CD's of NONMEM 7.2.0 to all users
with up-to-date licenses.
Some of the new features of NONMEM 7.2.0 include:
Dynamic Memory Allocation: No need to modify SIZES for unusually large
problems. Memory is automatically sized according to the
Leonid:
Could you please try this problem also with NONMEM 7.1.2 and nonmem
7.2.0
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics
ICON Development Solutions
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.com
Web: www.icondevsolutions.com
-Original
On 5/25/2011 9:43 PM, Bauer, Robert wrote:
Leonid:
Tom reminded me of our solution, shown below, which you have indeed
experienced with the more verbose error message in nm7.2. NONMEM's
classic design regarding MSF is that everything from the analysis is
captured has to be just right
...@metruminstitute.org]
Sent: Thursday, May 26, 2011 8:13 AM
To: Bauer, Robert
Cc: nmusers
Subject: Re: [NMusers] simulation problem or bug
Bob, Tom:
Thanks for sharing this explanation.
Users might actually be interested in conducting simulations from a
model that terminates with rounding errors
All:
I appreciate your efforts in working out the MPI installations for
various platforms, and disseminating the information to NONMEM users. I
will keep track of this information and add to the documentation to make
things clearer.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics
ICON
Nick and Andreas:
I do not have an appropriate windows network drive available for testing, but I
think I know what is going on. On Windows, when mpiexec is loaded and begins
executing through the call to nmmpi.bat, which is right after the statement
Starting MPI version of nonmem execution
Nick and Andreas:
I managed to get MPI work when running from a windows network drive by placing
-mapall on every command line to mpiexec. For example,
1:mpiexec -mapall -wdir %cd% -hosts 1 localhost 1 nonmem.exe %*
2-[nodes]:-mapall -wdir %cd%\worker{#-1} -hosts 1 localhost 1 nonmem.exe
If you are using NONMEM 7.2, IMAX can be increased by setting MAXFCN, by
inserting as the first executable line in the control stream file:
$SIZES MAXFCN=10
Keep in mind that as a 4 byte integer, MAXFCN cannot exceed 2**31=2147483648.
Robert J. Bauer, Ph.D.
Vice President,
Nick:
I recommend that you download and install a later version of the
gfortran compiler:
gcc-trunk-i686.tar.gz available at
ftp://nonmem.iconplc.com/Public/nonmem7/compilers/
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics
ICON Development Solutions
Tel: (215) 616-6428
Mob: (925)
-
From: Leonid Gibiansky [mailto:lgibian...@quantpharm.com]
Sent: Thursday, September 22, 2011 1:20 PM
To: Andrew Hooker
Cc: Rik Schoemaker; nmusers; Bauer, Robert
Subject: Re: [NMusers] Epsilon shrinkage and IWRES
Hi Andy,
Could you, please, provide an example of the model where Nonmem
The source code of LSODA is supplied in the ..\source directory of
NONMEM for NONMEM 7.1.2 and earlier. For NONMEM 7.2, the ..pr directory
has the PREDDPP source code available for viewing.
Typically when LSODA errors occur, is is corrected by modifying TOL.
However, sometimes LSODA errors
Dear Filip:
You may wish to try using NONMEM 7.2, which has an improved transition
process between SAEM and IMP EONLY=1. Add the extra option MAPITER=0:
$EST METHOD=IMP LIKE NITER=20 CTYPE=3 ISAMPLE=300 NOABORT EONLY=1
MAPITER=0
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics
ICON
Dear Nmusers:
A list of bugs for NONMEM 7.2 are located as file nm720_bug_list.pdf
ftp://nonmem.iconplc.com/Public/nonmem720/
An interim version of NONMEM will not be issued as these bugs have
work-arounds. These issues are being corrected for version NONMEM 7.3, to be
released in June 2013.
Sven:
You cannot easily bring in individual phi values externally. But, suitable
optimal phi values can be supplied to the BAYES method if you perform an ITS
analysis just before that. For example,
$EST METHOD=ITS INTERACTION NITER=0 (plus additional options)
$EST METHOD=BAYES INTERACTION
This is an erroneous statement by NONMEM given that convergence is not tested
when EONLY=1, even when CTYPE0, which you may ignore. It has been corrected
for NM7.2.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover,
ICON Development Solutions is pleased to announce workshops on NONMEM:
May 15-16 17, 2012
Howard Community College - Gateway Campus
6751 Columbia Gateway Drive (First Floor)
Columbia, Maryland 21046
Lab A / Classroom 3
May 15-16 - 2-Day Beginners Course
Introductory NONMEM PDx-Pop Hands-on
Paul:
Your g95 compiler is interfering with gfortran. Open a DOS window, then
set PATH=c:\program files\gfortran;%PATH%
set LIBRARY_PATH=
or wherever the proper gfortran460 is.
Then, rebuild your NONMEM, and test it.
This PATH statement must be re-specified each time you initiate a DOS
: www.iconplc.comhttp://www.iconplc.com/
From: owner-nmus...@globomaxnm.com [mailto:owner-nmus...@globomaxnm.com] On
Behalf Of Bauer, Robert
Sent: Friday, May 04, 2012 4:42 PM
To: prhut...@pharmacy.wisc.edu; nmusers@globomaxnm.com
Subject: RE: [NMusers] NM72 compilation
You should read the next several paragraphs in the guide, stating how you can
alternatively control NONMEM if the control keys do not work in certain
environments. This is done using the sig.exe program supplied with NONMEM.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.com
Web: www.iconplc.com
-Original Message-
From: Paolo Denti [mailto:paolo.de...@uct.ac.za]
Sent: Monday, May 28, 2012 4:54 AM
To: Bauer, Robert
Cc: Fidler,Matt,FORT WORTH,RD; nmusers
Subject: Re: SPAM: [NMusers
Thank you all for your suggestions. The more flexible OMEGA element entries
will not be easy to implement in NONMEM's structure, but I will look into it.
A couple of things need to be noted.
1) Regarding Matt's suggestions, to enter initial Omega elements in Cholesky
format in NM 7.2,
$OMEGA
In NONMEM 7.2, you may set:
$SIZES NO=500
or whatever size you need. This is described in nm720.pdf, section I.5
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
This is a test of the nmusers list-serve. Please discard.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.com
Web:
Pascal:
I cannot help regarding having all console messages sent to the proper files in
the PSN environment, but I can assist in avoiding your present NONMEM error.
If you insert at the beginning of the control stream file
$SIZES LIM1=??
and insert a large enough value for ??, then file buffer
Rebecca:
As there is no precise information about the error, it is possible that NONMEM
flagged it because a predictive value of F returned an Infinity or Not a
number. This could have been cause by a divide by 0 somewhere, or log of a
negative number, etc. You should check your code.
Filip:
The $TABLE results are obtained from a post-hoc assessment at the best fit
eta values (modal, EBE), regardless of the method used. To evaluate $TABLE
parameters at the conditional mean positions, you may select FNLETA=0.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON
, 2012 3:08 PM
To: Bauer, Robert; nmusers@globomaxnm.com
Subject: RE: NM7.2 SAEM with LIKE
Hi Bob,
Thanks for the clarification. Would you agree that the conditional means are
more trustworthy, in case there is a relevant difference? In my specific
dataset, there are 5 subjects for whom
) 286-0769
Email: robert.ba...@iconplc.com
Web: www.iconplc.comhttp://www.iconplc.com/
From: Manjunatha Shivaiah (BLR GSS) [mailto:manjunat...@nousinfo.com]
Sent: Thursday, August 23, 2012 7:03 AM
To: Bauer, Robert
Subject: FW: NONMEM 7.2 maximum parallelization nodes
Hello all:
I would like to remind everyone using nmusers to place appropriate subject
headers so that the archived e-mails can be easily searched for content.
For example, the most recent e-mail under
RE: [NMusers] single-patient AUC
(see e-mail trail below)
has nothing to do with
Yes, restarting NONMEM using the MSF system for a multi $EST methods problem is
sufficiently complex, that it is best to leave it to the user to decide what
needs to be done (that is, modifying the control stream file for the
appropriate re-start action), rather than trying to make an
The exit status=1 just means MPI is selected rather than FPI.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.com
Web:
Andreas:
NMTRAN has a a limit of accepting 2 digits for indexing etas, thetas, or
sigmas. This limitation has been removed for nm73, which is presently in beta
preparation.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Sven:
It would be mathematically incorrect to create simulated values using a Sobol
quasi-random method. You would obtain eta and eps values that are not truly
normally distributed. The Sobol technique is specifically designed to improve
the efficiency of Monte Carlo integration in
Nick:
The MU warnings are completely passive. I do not know what your control stream
would have. Any MU_X declarations?
Link errors are caused by your intel compiler may not be set up properly. It
has nothing to do with NMTRAN errors and warnings.
Robert J. Bauer, Ph.D.
Vice President,
there are
no MU_X declarations (see question 1).
The intel and g95 compilers works fine with all other problems I've
tried. Its only with this strange warning/error situation that there
seems to be a difference between them.
Nick
On 13/10/2012 1:35 p.m., Bauer, Robert wrote:
Nick:
The MU
: robert.ba...@iconplc.com
Web: www.iconplc.com
-Original Message-
From: Alison Boeckmann [mailto:alisonboeckm...@fastmail.fm]
Sent: Saturday, October 13, 2012 1:53 AM
To: Bauer, Robert; Nick Holford; nmusers
Subject: Re: [NMusers] What does MU_WARNING 24 mean?
Nick,
The subroutine CHECKMU
An alternative is to use the self-install versions of NONMEM for Windows using
gfortran that Bill and I have provided, which are located at:
ftp://nonmem.iconplc.com/Public/nonmem720
32 bit:
NONMEM720_gfortran460.exe
README_NONMEM720_gfortran460.exe
64 bit:
NONMEM720_64gfortran463.exe
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.com
Web: www.iconplc.comhttp://www.iconplc.com/
From: Kevin Dykstra [mailto:kevin.dyks...@qpharmetra.com]
Sent: Thursday, November 15, 2012 1:04 PM
To: Bauer, Robert
My receipt of Bill Denney's e-mail lagged somewhat. My explanation is similar
to his.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email:
Pascal:
There is one more consideration. If your model depends on the use of covariate
data, then during the numerical integration from time t1 to t2, where t1 and t2
are times of two contiguous records, which have values of the covariate c1 and
c2, respectively, NONMEM uses the covariate at
Dear NMusers:
ICON Development Solutions is presenting several workshops throughout the year.
Two workshops will be hosted in Ellicott City, Maryland, April 9-11, and
September 10-12. These are 3 day workshops, first two days are
beginner/intermediate workshops, and the third day is for new
Dear NMusers:
ICON Development Solutions is presenting several workshops throughout the year.
Two workshops will be hosted in Ellicott City, Maryland, April 9-11, and
September 10-12. These are 3 day workshops, first two days are
beginner/intermediate workshops, and the third day is for Monte
Actually, the ability to use SIGMA(1,1) on the right side has been around since
NONMEM VI release 2.
In the case of:
PROP=THETA(1)*F ; proportional part
ADD=THETA(2) ; additive part
SD=SQRT(PROP*PROP + ADD*ADD) ; Standard deviation using proportional
This could be transposed to:
Dear nmusers:
After a long and fruitful career in the pharmaceutical sciences and
pharmacometrics, William Bachman is retiring. While at ICON Bill has been the
primary technical support person for NONMEM and PDxPOP, as well as the primary
developer of PDx-POP. ICON will continue provide
The following is an advertisement for an omployment opportunity at ICON
Development Solutions.
Director, Pharmacometrics RD
Location- Anywhere in the US (office or home-based)
This is an exciting opportunity to work within a fast paced, busy environment
for a leading global provider of
There are several ways to do this.
Read about PARATYPE=2 in nm720.pdf
Or, instead of using psexec as the loader, use the native WINDOWS start
command. Here is an alternative fpiwini8.pnm using start:
$DEFAULTS
[nodes]=8
$GENERAL
NODES=[nodes] PARSE_TYPE=2 PARSE_NUM=34 TIMEOUTI=300
At present we do not have an easy install system for Linux, because of the
inconsistent environment. You will need to install from the CD.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Unfortunately there is a hard-coded limit for this option to 4 digits in
NTMRAN. I will raise to 12 digits for nm73
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
I made a quick check with nonmem 7.2, and I have no problem creating table
files even after two $TABLE statements that do not specify table files.
Maggie:
It appears specific to your control stream file. Please send me your control
stream and data file so I can determine if there is a bug in
Hello all:
New bug item in nm7.2.0, which will be resolved in nm73:
When a series of $TABLE statements without FILE= specification is followed by
$TABLE statements with FILE= specification, not all tables print out, and an
error is issued in the NONMEM report file: 0ERROR IN WRITING FILE :
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.commailto:robert.ba...@iconplc.com
Web: www.iconplc.comhttp://www.iconplc.com/
From: owner-nmus...@globomaxnm.com [mailto:owner-nmus...@globomaxnm.com] On
Behalf Of Bauer, Robert
Sent: Tuesday, July 09, 2013 9:17 PM
To: Ron
Dear NMusers:
ICON Development Solutions is presenting a 4-day workshop which will be hosted
in Howard County, Maryland, September 10-13.
First two days are beginner/intermediate NONMEM / PDxPoP workshops, and the
third day is for Monte Carlo Expectation/Maximization and full Bayesian
If you mean the iteration number of the estimation, this is accessed through
ITER_REPORT by inserting:
$PK
FIRST
USE NMBAYES_INT, ONLY: ITER_REPORT
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel:
Dear Beta Testers:
An eighth release of the beta version of NONMEM 7.3.0 is now available
for testing.
The complete list of changes from NONMEM 7.2 to NONMEM 7.3, with references to
their descriptions in ..\guides\nm730.pdf, is given in readme_nm730.pdf,
readme_nm730.txt, and
Dear NMusers:
ICON Development Solutions is presenting a 4-day workshop which will be hosted
in Howard County, Maryland, April 8-11.
First two days are beginner/intermediate NONMEM / PDxPoP workshop
The third day is for Monte Carlo Expectation/Maximization and full Bayesian
analyses. In
Dear NMUSERS:
NONMEM 7.3.0 is now available.
Please visit the site
ftp://nonmem.iconplc.com/Public/nonmem730
and view readme_730.pdf. This file lists new features of NONMEM 7.3.0 and bug
fixes. The guide nm730.pdf contains a more detailed description of the new
features.
Lisa Wilhelm has
Dear nmusers:
Continuing the workshop topic, ICON and QuantPharm will team up again to
present two Workshop series that will combine NONMEM course with the course on
TMDD modeling:
April 8-11, 2014, Maryland, US (near Baltimore):
Days 1-2: Beginner/intermediate NONMEM / PDxPoP workshops
Douglas:
Thank you for sharing your code and data with me. When PRED, WRES, or RES are
requested, NONMEM calls the weighted residual routine PRRES, which calculates
PRED, WRES, and RES together.
Your problem has a large number of data per subject (up to 1544) , and
evaluating WRES requires
Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.com
Web: www.iconplc.com
-Original Message-
From: Bob Leary [mailto:bob.le...@certara.com]
Sent: Wednesday, January 29, 2014 7:03 PM
To: Bauer, Robert; Eleveld, DJ; nmusers@globomaxnm.com
I concur with Ken's statement, and I also prefer to use MATRIX=R as the first
choice for covariance assessment. On occasion, MATRIX=S can be used if there
are numerical difficulties in assessing the R matrix, and if there are enough
subjects relative to the dimension size (number of total
This does leave out the ability to model a power term, theta(2). Fortunately,
if there are fewer than 4 thetas that are not mu-referenced, usually the EM
algorithms can still solve the problem, but with less efficiency. The
parameterization I recommend for this particular problem, and include
Testing nmusers, please ignore.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.commailto:robert.ba...@iconplc.com
Web:
Pavel:
Try using the latest nonmem 7.3 version if you can, and see if this still
occurs. Improvements are continually being made. Try also the following after
an SAEM step:
$EST METHOD=IMP EONLY=1 MAPITER=0 (...plus other options as neded)
Or
$EST METHOD=IMPMAP EONLY=1 (...plus other options
Bob:
Yes, when I was developing the code for using the Sobol method in NONMEM, I
too found that using the t-distribution with Sobol/Quasi normal method resulted
in biased assessments of the objective function, when I used the unmodified
standard technique of generating t-samples. So, I
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.commailto:robert.ba...@iconplc.com
Web: www.iconplc.comhttp://www.iconplc.com/
From: Bob Leary [mailto:bob.le...@certara.com]
Sent: Monday, May 19, 2014 2:36 PM
To: Bauer, Robert; nmusers
Dear NONMEM Users:
1) NONMEM 7.3 has been available since December 2013, and many of you have
been using the latest features that it has to offer. Minor bugs have been
reported, for which there are work-arounds. These are published at the
following location:
Make sure ALL your $OMEGA records are placed just before $PK
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.com
Web: www.iconplc.com
The following location might be more up-to-date:
ftp://nonmem.iconplc.com/Public/nonmem/tips
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email:
Paolo:
MCETA=5 is a small number of random samples of etas to test, and when you
parallelize, your random seed pattern changes, unless, you include $EST
RANMETHOD=P. Yes, sometimes a worse result follows if a lower OBJ is chosen,
yet those etas that gave a lower OBJ offered a worse path than
Sheng:
This seems like a rather arbitrary occurrence. If you would supply me with
your problem (control stream file and data set), I would be happy to determine
if there is a bug. Please let me know what version of NONMEM you are using,
compiler, and operating system.
Robert J. Bauer, Ph.D.
I have forwarded your query to Andy Hooker
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.commailto:robert.ba...@iconplc.com
Web:
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.commailto:robert.ba...@iconplc.com
Web: www.iconplc.comhttp://www.iconplc.com/
br /br /
The new gfortran compiler appears to report any IEEE signals. Your particular
control stream seems to cause a dive-by zero that does not impact your results.
Nonetheless, browse through your control stream and very that variables that
you divide by are never zero, such as your residual weight
This is a test.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.commailto:robert.ba...@iconplc.com
Web:
This is a test
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215) 616-6428
Mob: (925) 286-0769
Email: robert.ba...@iconplc.commailto:robert.ba...@iconplc.com
Web:
Dear nmusers members:
ICON will be hosting a booth at the ASCPT 2015 meeting in New Orleans. Meet us
at booth #111 to discuss your Early Phase drug development requirements,
population PK/PD analysis and NONMEM/PDx-Pop. ICON delivers expertise in the
design, study conduct and interpretation
Dennis:
The latest gfortran compiler sends these signals, but they do not impact
results. I am still looking for a way to turn these messages off.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel: (215)
Dear nmusers Members:
ICON in association with Keio University is pleased to present
beginning/intermediate and advanced workshops on PDx-Pop and NONMEM for PK/PD
population analysis, in Yokohama Japan, on November 3-5, 2015. Prof. Yusuke
Tanigawara of Keio University has graciously agreed to
Dear nmusers members:
ICON will be hosting a booth at the ASCPT 2015 meeting in New Orleans. Meet us
at booth #111 to discuss your Early Phase drug development requirements,
population PK/PD analysis and NONMEM/PDx-Pop. ICON delivers expertise in the
design, study conduct and interpretation
All requests for removal from or addition to the nmusers list must be submitted
to nmusers-requ...@globomaxnm.commailto:nmusers-requ...@globomaxnm.com
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics, RD
ICON Development Solutions
7740 Milestone Parkway
Suite 150
Hanover, MD 21076
Tel:
[mailto:ms...@nuventra.com]
Sent: Tuesday, February 24, 2015 6:26 PM
To: nmusers@globomaxnm.com; Bauer, Robert
Subject: Error in parallel nonmem - related to file size?
Dear NMusers (and Bob).
I'm getting an error running parallel NONMEM, with MPI. It seems that
perhaps the file size
Chuanpu:
Matt Hutmacher and I are working on a plug-in ($SUBROUTINES OTHER=) function to
deal with bivariate normal cumulative distributions. We hope to upload this on
the ICON nonmem site by end of the week.
Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics RD
ICON Early Phase
Office:
Bill:
I have just recently tested this with nonmem 7.3, and the fort.50 and fort.51
files show up on the worker directories, so I am not sure if something is going
on in your environment. If you can describe your OS, and send your pnm file,
I'll try to figure out what it going on. By the
Xinting Wang:
Generating each sample in SAEM is computationally more expensive in SAEM, and
so ISAMPLE of 2-10 is generally done. You may try increasing ISAMPLE=3 or 5 in
SAEM. Furthermore, convergence testing is only a guide, Review the .ext file
and be satisfied by visual inspection that
1 - 100 of 261 matches
Mail list logo
