# RE: [NMusers] eigenvalues

From: Matt Hutmacher <matt.hutmacher_at_a2pg.com>
Date: Fri, 6 Nov 2015 15:37:53 -0500

Hello Pavel,

Could you share the COV matrices from Monolix and NONMEM? I have an =
idea of what could be going on, but it would be good to see the matrices =
to check if the hypothesis makes sense.

Best,
Matt

From: owner-nmusers_at_globomaxnm.com [mailto:owner-nmusers_at_globomaxnm.com] =
On Behalf Of Jeroen Elassaiss-Schaap
Sent: Friday, November 06, 2015 11:34
To: Pavel Belo
Cc: “nmusers_at_globomaxnm.com”
Subject: Re: [NMusers] eigenvalues

Hi Pavel,

For starters, it is simple to calculate using R:
mymat<-abs(matrix(rnorm(25^2),ncol=25))
mymat <- mymat /max(mymat)
#replace mymat with your nonmem \$cov matrix
eigenval<-eigen(mymat,symm=T)\$values # should be similar to nonmem =
reported
cn<-max(eigenval)/min(eigenval)
eigenval<-eigen(mymat[1:10,1:10],symm=T)\$values
cn1<-max(eigenval)/min(eigenval) # could be compared to the "PK" =
parameters ratio from monolix

Assuming a 25x25 covariance matrix, and theta in 1:10. You will need to =
do some rearrangement of the cells to isolate the off-diagonal elements =
of \$OMEGA, but with this approach you can compare apples by apples. =
Until you have done that you will not know whether the platforms provide =
different results or similar wrt the condition number.

The difference in behavior with respect to objective function impact is =
puzzling, assuming you refer to SAEM estimation in Nonmem. My advice =
here would be to focus on (visual) predictive checks, and compare how =
well the two platforms perform on that aspect.

Hope this helps,
Jeroen

--

http://pd-value.com
jeroen_at_pd-value.com
_at_PD_value
+31 6 23118438
-- More value out of your data!

Received on Fri Nov 06 2015 - 15:37:53 EST

This archive was generated by hypermail 2.3.0 : Fri Sep 27 2019 - 16:46:58 EDT