NONMEM Users Network Archive

Hosted by Cognigen

RE: Statistical power computation based on the wald test

From: Ken Kowalski <kgkowalski58>
Date: Mon, 22 Mar 2021 10:34:35 -0400

Hi Ibtihel,


I think you are probably asking for covariance matrix of the parameter
estimates. This should automatically be outputted as the .cov file assuming
that the $COV step runs successfully. Note that since NONMEM minimizes a
function related to -2LL, the Hessian (R matrix) in NONMEM is equivalent to
Fisher's Informaton matrix. I know you can print the R matrix in the NONMEM
output and I assume this can also be outputted to a file.perhaps others
might know


I'll leave it for you to decide whether you really want to perform power
calculations say to design/justify a sample size to detect the covariate
effect using a Wald-based test as opposed to performing simulations and
relying on a likelihood ratio test.




Kenneth G. Kowalski

Kowalski PMetrics Consulting, LLC

Email: <mailto:kgkowalski58

Cell: 248-207-5082




From: owner-nmusers
Behalf Of Hammami, Ibtihel /FR
Sent: Monday, March 22, 2021 9:22 AM
To: nmusers
Subject: [NMusers] Statistical power computation based on the wald test



We would like to compute a covariate inclusion statistical power based on
the Wald test and using SE given by the fisher information matrix.

Is there any method to implement this directly in NONMEM or is there at
least a way to output the Fisher Information matrix in NONMEM?


Thank you.


This email has been checked for viruses by Avast antivirus software.

Received on Mon Mar 22 2021 - 10:34:35 EDT

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: