NONMEM Users Network Archive

Hosted by Cognigen

[NMusers] RE: Mixture model with logistic regression

From: Bob Leary <>
Date: Sat, 20 Feb 2016 14:24:36 +0000

OK - but there's no inherent reason why LIKE could not work with NP - the u=
nderlying theoretical NP algorithm is totally agnostic to where the likel=
ihoods come from or what type of observation is being used. Certainly in=
 Phoenix NLME this works. Not sure about USC*PACK. If you can get NONMEM =
to output a table of posthocs and correpsonding likelihoods, then this is e=
asy to try in MATLAB (I would be happy to share a simple m-file that
implements Bradley Bells excellent primal dual algorithm for this, or simp=
ly run it and pass back the results).
From: Mark Sale []
Sent: Saturday, February 20, 2016 8:07 AM
To: Bob Leary;
Subject: Re: Mixture model with logistic regression


That certainly makes sense, but that options seems to not be available in N=
ONMEM, using LIKE seems to require using FOCE LAPLACE

      This is designed mainly, but not exclusively, for use with non-
      continuous observed responses ("odd-type data"). Indicates that
      Y (with NM-TRAN abbreviated code) or F (with a user-supplied PRED
      or ERROR code) will be set to a (conditional) likelihood. Upon
      simulation it will be ignored, and the DV data item will be set
      directly to the simulated value in abbreviated or user code.
      Also etas, if any, are understood to be population etas. Epsilon
      variables and the $SIGMA record may not be used. The L2 data
      item may not be used. The CONTR and CCONTR options of the $SUB-
      ROUTINES record may not be used. NONMEM cannot obtain the ini-
      tial estimate for omega. If the data are population, and MAXE-
      VALS=0 is not coded, then METHOD=1 LAPLACE is required. Compare
      with PREDICTION option.

Mark Sale M.D.
Vice President, Modeling and Simulation
Nuventra, Inc.
2525 Meridian Parkway, Suite 280
Research Triangle Park, NC 27713
Office (919)-973-0383<UrlBlockedError.aspx><>

Empower your Pipeline

CONFIDENTIALITY NOTICE The information in this transmittal (including attac=
hments, if any) may be privileged and confidential and is intended only for=
 the recipient(s) listed above. Any review, use, disclosure, distribution o=
r copying of this transmittal, in any form, is prohibited except by or on b=
ehalf of the intended recipient(s). If you have received this transmittal i=
n error, please notify me immediately by reply email and destroy all copies=
 of the transmittal.

From: Bob Leary <>
Sent: Saturday, February 20, 2016 8:45 AM
To: Mark Sale;
Subject: RE: Mixture model with logistic regression

This sounds like a good case for a nonparametric method - if you use the o=
ne in NONMEM, you might try
expanding Omega to counter shrinkage. The versions in USC*PACK and PHOENI=
X NLME optimize over
both support point positions and probabilities, so this is not necessary wi=
th those methods.
From: [] on behalf=
 of Mark Sale []
Sent: Friday, February 19, 2016 4:30 PM
Subject: [NMusers] Mixture model with logistic regression

Has anyone every tried to use a mixture model with logistic regression? I h=
ave data on a AE in several hundred patients, measured multiple times (10-2=
0 times per patient). Examining the data it is clear that, independent of =
drug concentration, there is very wide distribution of this AE, 68% of the =
patients never have the AE, 25% have it about 20% of the time and the rest =
have it pretty much continuously, regardless of drug concentration. (in or=
dinary logistic regression, just glm in R, there is also a nice concentrati=
on effect on the AE in addition). Running the usual logistic model, not s=
urprisingly, I get a really big ETA on the intercept, with 68% of the peopl=
e having ETA small negative, 25% ETA ~ 1 and 7% ETA ~ 10. No covariates see=
m particularly predictive of the post hoc ETA. I thought I could use a mix=
ture model, with 3 modes, but it refused to do that, giving me essentially =
0% in the 2nd and 3rd distribution, still with the really large OMEGA for t=
he intercept. Even when I FIX the OMEGA to a reasonable number, I still ge=
t essentially no one in the 2nd and 3rd distribution. I tried fixing the f=
raction in the 2nd and 3rd distribution (and OMEGA), and it still gave me a=
 very small difference in the intercept for the 2nd and 3rd populations.

Is there an issue with using mixture models with logistic regression? I'm j=
ust using FOCE, Laplacian, without interaction, and LIKE.

Any ideas?


Mark Sale M.D.
Vice President, Modeling and Simulation
Nuventra, Inc.
2525 Meridian Parkway, Suite 280
Research Triangle Park, NC 27713
Office (919)-973-0383<UrlBlockedError.aspx><>

NOTICE: The information contained in this electronic mail message is intend=
ed only for the personal and confidential use of the designated recipient(s=
) named above. This message may be an attorney-client communication, may be=
 protected by the work product doctrine, and may be subject to a protective=
 order. As such, this message is privileged and confidential. If the reader=
 of this message is not the intended recipient or an agent responsible for =
delivering it to the intended recipient, you are hereby notified that you h=
ave received this message in error and that any review, dissemination, dist=
ribution, or copying of this message is strictly prohibited. If you have re=
ceived this communication in error, please notify us immediately by telepho=
ne and e-mail and destroy any and all copies of this message in your posses=
sion (whether hard copies or electronically stored copies). Thank you.


Received on Sat Feb 20 2016 - 09:24:36 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: