calcnpde {npde}R Documentation

Compute the prediction distribution errors in one subject

Description

For each subject, this function computes the prediction errors and returns the decorrelated simulated and observed data.

Usage

calcnpde(isuj, msuj, matsim, nrep, verbose)

Arguments

isuj subject ID
msuj observed data for the subject, a dataframe containing 3 columns (id=patient ID, xobs=independent variable (X), yobs=dependent variable (Y)
matsim simulated data for the subject, a dataframe containing 4 columns (idsim=patient ID, irsim=integer identifying the replicate; xsim=independent variable (X), ysim=dependent variable (Y)
nrep number of replications
verbose a boolean (T if messages are to be printed as each subject is processed, F otherwise)

Value

xerr an integer code to keep track of errors during the computation; after a successful computation the value of xerr should be 0. A value of 1 or 2 signals errors during the computation.
pde the vector of prediction distribution errors for the subject
ydsim the vector of simulated dependent variable (Y) for the subject, decorrelated
ydobs the vector of observed dependent variable (Y) for the subject, decorrelated

Author(s)

Emmanuelle Comets <emmanuelle.comets@bichat.inserm.fr>

References

K. Brendel, E. Comets, C. Laffont, C. Laveille, and F. MentrĂ©. Metrics for external model evaluation with an application to the population pharmacokinetics of gliclazide. Pharmaceutical Research, 23:2036–49, 2006.

See Also

npde, autonpde


[Package npde version 1.2 Index]