NONMEM Users Network Archive

Hosted by Cognigen

RE: Standard errors of estimates for strictly positive parameters

From: pascal.girard
Date: Wed, 11 Feb 2015 17:29:45 +0000

Dear Aziz,

NM does not return the asymptotic SE of THETA(1) in model 1 on the log-scal=
e. So I would use model 2.

With best regards / Mit freundlichen Grüßen / Cordialement


From: owner-nmusers
 Behalf Of Chaouch Aziz
Sent: 11 February 2015 17:22
To: nmusers
Subject: [NMusers] Standard errors of estimates for strictly positive param=


I'm interested in generating samples from the asymptotic sampling distribut=
ion of population parameter estimates from a published PKPOP model fitted w=
ith NONMEM. By definition, parameter estimates are asymptotically (multivar=
iate) normally distributed (unconstrained optimization) with mean M and cov=
ariance C, where M is the vector of parameter estimates and C is the covari=
ance matrix of estimates (returned by $COV and available in the lst file).

Consider the 2 models below:

Model 1:



Model 2:



It is clear that model 1 and model 2 will provide exactly the same fit. How=
ever, although in both cases the standard error of estimates (SE) will refe=
r to THETA(1), the asymptotic sampling distribution of TVCL will be normal =
in model 1 while it will be lognormal in model 2. Therefore if one is inter=
ested in generating random samples from the asymptotic distribution of TVCL=
, some of these samples might be negative in model 1 while they'll remain n=
icely positive in model 2. The same would happen with bounds of (asymptotic=
) confidence intervals: in model 1 the lower bound of a 95% confidence inte=
rval for TVCL might be negative (unrealistic) while it would remain positiv=
e in model 2.

This has obviously no impact for point estimates or even confidence interva=
ls constructed via non-parametric bootstrap since boundary constraints can =
be placed on parameters in NONMEM. But what if one is interested in the asy=
mptotic covariance matrix of estimates returned by $COV? The asymptotic sam=
pling distribution of parameter estimates is (multivariate) normal only if =
the optimization is unconstrained! Doesn't this then speak in favour of mod=
el 2 over model 1? Or does NONMEM take care of it and returns the asymptoti=
c SE of THETA(1) in model 1 on the log-scale (when boundary constraints are=
 placed on the parameter)?


Aziz Chaouch
This message and any attachment are confidential and may be privileged or o=
therwise protected from disclosure. If you are not the intended recipient, =
you must not copy this message or attachment or disclose the contents to an=
y other person. If you have received this transmission in error, please not=
ify the sender immediately and delete the message and any attachment from y=
our system. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do n=
ot accept liability for any omissions or errors in this message which may a=
rise as a result of E-Mail-transmission or for damages resulting from any u=
nauthorized changes of the content of this message and any attachment there=
to. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not guara=
ntee that this message is free of viruses and does not accept liability for=
 any damages caused by any virus transmitted therewith.

Click to access the German, French, Sp=
anish and Portuguese versions of this disclaimer.

Received on Wed Feb 11 2015 - 12:29:45 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: