npde (version 3.5)

npde.cens.method: Method used to handle censored data

Description

Specifies the method used to handle censored data (data below the limit of quantification LOQ

Value

This is not a function and does not have a return value, this is a statistical method.

More details can be found in the PDF documentation.

Details

Several methods are available to handle censored data.

omit

pd and npde for censored data will be set to NA

cdf

for an observation ycens_ij under the LOQ, a pd_ij will be imputed in the uniform distribution [0-pLOQ_ij] where pLOQ_ij is the probability that y_ij is below LOQ, according to the model; the predictive distribution will then be used to obtain a corresponding y*_ij. This is also performed for all simulated data, and the npde are then computed on the completed dataset containing the observed y_ij for the uncensored data and the y*_ij imputed for the censored data. This method is the default.

ipred

an observation ycens_ij is replaced by the individual prediction according to the model (ipred, which must be present in the dataset). Simulated data are left untouched.

ppred

an observation ycens_ij is replaced by the population prediction according to the model. Simulated data are left untouched.

loq

an observation ycens_ij is replaced by the value of the LOQ. Simulated data are left untouched.

References

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

Examples

if (FALSE) {
# You need to have gridExtra installed to successfully run this example
if(requireNamespace("gridExtra", quietly=TRUE)) {
  data(warfarin)
  data(simwarfarinCov)
  wcov<-autonpde(namobs=warfarin,namsim=simwarfarinCov, iid=1,ix=2,iy=4,icov=c(3,6:8),
  namsav="warfCov", units=list(x="hr",y="mg/L", covariates=c("mg","kg","-","yr")))
  
  # Diagnostic plots for warfarin with a covariate model
  plot(wcov)
  
  # Covariate plots
  xwt.scatt<-plot(wcov, plot.type="x.scatter", covsplit=TRUE, which.cov="wt")
  xwt.qqplot<-plot(wcov, plot.type="qqplot", covsplit=TRUE, which.cov="wt")
  xwt.box<-plot(wcov, plot.type="covariates", which.cov="wt")
  xsex.scatt<-plot(wcov, plot.type="x.scatter", covsplit=TRUE, which.cov="sex")
  xsex.qqplot<-plot(wcov, plot.type="qqplot", covsplit=TRUE, which.cov="sex")
  xsex.box<-plot(wcov, plot.type="covariates", which.cov="sex")
  
  # Transforming the reference profile for npd, compared to a VPC plot
  plot.tnpd<-plot(wcov, plot.type="x.scatter", ref.prof=list(id=2), 
  main="tnpd with reference profile ID=2")
  plot.vpc<-plot(wcov, plot.type="vpc", main="VPC")
  gridExtra::grid.arrange(grobs=list(plot.tnpd, plot.vpc), nrow=1, ncol=2)
  }
}