We illustrate the use of the package with two simulated datasets, one under the true model and one with different parameter values, to show how npde can be used to evaluate models.
Npde nonmem software#
Simulations need to be performed before hand, using for example the software used for model estimation. or plots of pd and npde versus time and predictions are useful to highlight. Under the null hypothesis that the model under scrutiny describes the validation dataset, npde should follow the standard normal distribution. NPDE thus appear as a good tool to evaluate population models, with or without covariates. They can be viewed as a good alternative way of evaluation by looking at MC simulated predictions. NPDE do not depend on an approximation of the model and have good statistical properties. Diagnostic graphs are produced for npd, and npde are used in the tests as their distribution takes into account the correlation. The use of NPDE over WRES is recommended for model evaluation. npde is an interactive function whereas autonpde takes all required input as arguments.
Npde nonmem full#
npde take into account the full predictive distribution of each individual observation and handle multiple observations within subjects. Both functions compute the normalised prediction distribution errors (and/or prediction discrepancies) in the same way.
First make a copy of your final model file (if your model is called run01.mod make it run01npde. In this paper, we describe an add-on package for the open source statistical package R, designed to compute npde. The way I learned to perform an NPDE analysis could be considered old-school but it worked through different versions of NONMEM and R, and this is why I prefer it. Recently, normalised prediction distribution errors (npde) have been proposed as a model evaluation tool. Pharmacokinetic/pharmacodynamic data are often analysed using nonlinear mixed-effect models, and model evaluation should be an important part of the analysis.
Computing normalised prediction distribution errors to evaluate nonlinear mixed-effect models: The npde add-on package for R.
0 kommentar(er)
