[NMusers] model of error, linearisation, additive error, and VPC

From: Pavel Belo <nonmem_at_optonline.net>
Date: Fri, 29 Jan 2016 16:47:25 -0500 (EST)





Hello NONMEM Users,



When I tried to print log-scaled VPC, there was an error message about
negative
values.  It can be caused by an additive error and/or linearization of=
 
the
error model when NONMEM transforms 
Y  = F*DEXP(ERR(1)*SD1)  into Y  =
F+ F*ERR(1)*SD1.  The error is inflated at small concentrations a=
nd
removing the additive term
is not an option in my case.  
 
Is there an easy way to
solve it?  It can be something like Y  = F**GAMMA*DEXP(ERR(1)*S=
D1)  or
Y  = F+ F*ERR(1)*SD1 + F**GAMMA*ERR(2)*SD2, where GAMMA<1.=
 


 


Thanks,


Pavel





Received on Fri Jan 29 2016 - 16:47:25 EST

This archive was generated by hypermail 2.3.0 : Fri Sep 27 2019 - 16:48:45 EDT