log additive error model

Phoenix Certara Forums Modelling and Simulation
1

Hello Everyone,

I am trying to fit a PK PD model wherein I am freezing the PK part.

For the PK and PD part I am using a log additive error model .

When the results are spit out, I usually get a log transformed values for the fit.

For some reason I don’t see my final fit in a ln transformed manner. Is this a bug ? or something is amiss in the code below:

test(){
deriv(A1 = - Vmax * C / (Km + C) - Cl2 * (C - C2)- ClC)
deriv(A2 = Cl2 * (C - C2))
deriv(Ce = Ke0(C - Ce))
dosepoint(A1)
C = A1 / V
C2 = A2 / V2

E = E0 + Emax * Ce / (EC50 + Ce)
error(CEps = 0.222147 )
observe(CObs = C *exp( CEps))

error(EEps = 1)

observe(EObs = E *exp(EEps))
stparm(V = tvV )
stparm(V2 = tvV2 )
stparm(Km = tvKm )
stparm(Vmax = tvVmax )
stparm(Cl = tvCl )
stparm(Cl2 = tvCl2 )
stparm(Ke0 = tvKe0 )
stparm(EC50 = tvEC50 )
stparm(E0 = tvE0)

stparm(Emax = tvEmax)

fixef(tvV(freeze) = c(0, 0.0630111, ))
fixef(tvV2(freeze) = c(0, 0.0326499, ))
fixef(tvKm(freeze) = c(0, 0.647138, ))
fixef(tvVmax(freeze) = c(0, 0.118141, ))
fixef(tvCl2(freeze) = c(0, 0.0207028, ))
fixef(tvCl(freeze) = c(0, 0.007726, ))

fixef(tvKe0 = c(0, , ))
fixef(tvEC50 = c(0, , ))
fixef(tvE0 = c(0, , ))
fixef(tvEmax = c(0, , ))

}

Thank you

2

It is because you have 2 observe statements. You will see the log-transformed values in the residual table only if you have one observe statement only.

best

Serge