From:Paul Hutson
Subject:[NMusers] Sensitivity analysis
Date:Thu, 05 Sep 2002 13:50:40 -0500
Hello, everyone.
Has anyone suggestions on how to perform a sensitivity analysis (if one is
possible) when using MAXEVALS=0 to perform a "Bayesian" fit of new data
using prior reports of PK parameters as the respective initial estimates of
$THETA ? Other than tweaking the OMEGAs to see the impact on new THETA
estimates, there doesn't seem to be much left to alter.
It also would seem to me that using the minimization of the objective
function for a MAXEVAL=0 run would potentially be misleading in
interpreting the optimality of the resulting THETAs. Could use the
appearance of the residual plots, I suppose...
What advice do you have?
Thanks
Paul
Paul Hutson, Pharm.D.
Associate Professor (CHS)
UW School of Pharmacy
777 Highland Avenue
Madison, WI 53705-2222
Tel: (608) 263-2496
FAX: (608) 265-5421
Pager: (608) 265-7000, #7856