Hi Niyi,
You may also try new metrics like the npde described at :
( http://www.page-meeting.org/?abstract=1120)
"Normalised prediction distribution errors in R: the npde library "
Emmanuelle Comets , Karl Brendel and France Mentré
The library is downloadable over the internet.
Bests,
Samer
-------- Message d'origine--------
Quoted reply history
De: [EMAIL PROTECTED] de la part de NIYI ADEDOKUN
Date: lun. 09/07/2007 18:10
À: [email protected]
Objet : [NMusers] COMPUTING COVARIANCE OF PREDICTION ERRORS IN NONMEM
Dear NMUsers,
Sometimes, it is desirable to compute an estimate of the covariance of 2
prediction errors in the same individual in order to estimate the variance of
the sum of prediction errors in that individual.
This estimate has been suggested by Vozeh et al. (1990) to be in the form:
ole0.bmp
http://by114fd.bay114.hotmail.msn.com/cgi-bin/getmsg?&msg=A0E3E567-5045-49A7-A890-C036EA33107C&start=0&len=139777&curmbox=00000000-0000-0000-0000-000000000001&a=11899ac4de63cb63ade73ddc25f986007f0c52e500b22c80fb4327221aad9394&mimepart=5
where where covN(PEi,PEi') is the covariance of the prediction errors i and i'
while E is the expectation of the enclosed expression; g and h are partial
derivatives of the vector of PK parameters with respect to the ç and å
respectively, m is the number of etas in the model etc.)
My question is: how can this be coded/outputted from a NONMEM run.
Thanks for your help.
Jo