NONMEM Users Network Archive

Hosted by Cognigen

RE: NM7 Importance Sampling: Objective function goes to zero, but results look reasonable

From: De Ridder, Filip [JRDBE] <FDRIDDER>
Date: Thu, 8 Dec 2011 11:32:27 +0100

Hi Bob,

 

NONMEM 7.2 behaves as expected in this regard!

 

I do have another question. I find a very large numerical difference
between the final SAEMOBJ and the IMP OBJ. For example, in SAEMOBJ
-21378, IMP OBJ=-1120. The latter seems very reasonable as it is
similar to the OBJ value of a terminated FOCE run.

Also, with the same model and data, changing the initial values results
in a rather different SAEMOBJ (-21103), but finally ending up with the
same IMP OBJ (and parameter estimates).

 

Is this large difference worrisome? In the few SAEM applications I have
seen in Monolix, the difference between these the SAEMOBJ and the IMP
OBJ is not that large.

 

Kind regards,

 

 

Filip De Ridder

Janssen Research & Development, Beerse, Belgium.

 

 

From: Bauer, Robert [mailto:Robert.Bauer
Sent: Monday, 10 October 2011 17:30
To: De Ridder, Filip [JRDBE]; nmusers
Subject: RE: [NMusers] NM7 Importance Sampling: Objective function goes
to zero, but results look reasonable

 

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 Development Solutions

Tel: (215) 616-6428

Mob: (925) 286-0769

Email: Robert.Bauer

Web: www.icondevsolutions.com <outbind://220/www.icondevsolutions.com>

 

 

 

________________________________

From: owner-nmusers
On Behalf Of De Ridder, Filip [JRDBE]
Sent: Monday, October 10, 2011 4:41 AM
To: nmusers
Subject: [NMusers] NM7 Importance Sampling: Objective function goes to
zero, but results look reasonable

Dear NMUsers,

 

When using SAEM/IMP in NM7.1., I get a peculiar results from IMP. After
a successful SAEM-run resulting in reasonable point estimates, the
objective function is first declared as 0.00000 and later as "NaN" at
the final iteration. I do get standard-errors and these look
reasonable, but I am not sure how trustworthy they are given the NaN
OBJ.

This is the $EST call:

 

$EST METHOD=SAEM LAPLACE LIKE NBURN=1000 NITER=2000 PRINT=10 =
CTYPE=3
CINTERVAL=10 NOABORT

$EST METHOD=IMP LIKE NITER=20 CTYPE=3 ISAMPLE=300 NOABORT =
EONLY=1

 

FOCE/Laplace terminates due to rounding errors, but gives reasonable
point estimates close to the SAEM result.

 

Below is a selection of the search.

 

Kind regards,

 

Filip De Ridder

Janssen Research & Development, Beerse, Belgium.

 

 #METH: Stochastic Approximation Expectation-Maximization

EM/BAYES SETUP

THETAS THAT ARE MU MODELED:

   1 2 3 4 5 6 7 8 9 11 12

THETAS THAT ARE SIGMA-LIKE:

 

 

MONITORING OF SEARCH:

 

Stochastic/Burn-in Mode

iteration -1000 SAEMOBJ= 26003.614680275896

 iteration -990 SAEMOBJ= -1092.1038437297473

 ...

iteration -670 SAEMOBJ= -3257.6199048158469

 iteration -660 SAEMOBJ= -3454.3544417057101

 Convergence achieved

Reduced Stochastic/Accumulation Mode

iteration 0 SAEMOBJ= -3365.8558463970126

 iteration 10 SAEMOBJ= -3739.6849187475736

 ...

iteration 1960 SAEMOBJ= -3823.2755123680026

 iteration 1970 SAEMOBJ= -3823.2896277883360

 iteration 1980 SAEMOBJ= -3823.4174894810230

 iteration 1990 SAEMOBJ= -3823.3225200406459

 iteration 2000 SAEMOBJ= -3823.2969976068907

 Elapsed estimation time in seconds: 2862.05

 

#TERM:

STOCHASTIC PORTION COMPLETED

REDUCED STOCHASTIC PORTION COMPLETED

 

 

#METH: Objective Function Evaluation by Importance Sampling

EM/BAYES SETUP

THETAS THAT ARE MU MODELED:

   1 2 3 4 5 6 7 8 9 11 12

THETAS THAT ARE SIGMA-LIKE:

 

 

MONITORING OF SEARCH:

 

iteration 0 OBJ= 0.0000000000000000

 iteration 10 OBJ= NaN

Elapsed estimation time in seconds: 1472.26

iteration 20 OBJ= NaN

 

#TERM:

OPTIMIZATION NOT TESTED

 

 

 

 

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
 

Received on Thu Dec 08 2011 - 05:32:27 EST

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.