computePredictions
[Monolix] Compute predictions from the structural model
Compute predicted observation model values on observation times for each individual in a set of individuals. By default, the predictions
are computed for all individuals. If the parameter individualIds
is specified, the individualParameters
must contain only
the parameters for those individuals. That is, the number of rows in individualParameters
must be the same as the length of individualIds
.
Usage
computePredictions(individualParameters, individualIds = NULL)
Arguments
- individualParameters
Individual parameter values for each parameter present in the model, either for all individuals or for the set of individuals specified in
individualIds
". This should be a data.frame with a column for individual id and a column for each parameter.- individualIds
[optional] vector Ids of the individuals for which observation models should be computed. By default, all the individuals are used.
Value
A list of predictions, where each prediction is a vector giving the computed prediction at observation times for each individual
Details
For each prediction, all individual values are returned as a single vector. Thus the number of observations per individual can be used to separate the predictions per individual.
See also
getIndividualParameterModel
to get the individual parameter model used for the prediction getEstimatedIndividualParameters
to get individual parameters to use in the prediction
Examples
initializeLixoftConnectors("monolix")
project_file <- file.path(getDemoPath(), "1.creating_and_using_models", "1.1.libraries_of_models", "theophylline_project.mlxtran")
loadProject(project_file)
runScenario()
ids <- c(1,4)
individualParamsForAllIndiv <- getEstimatedIndividualParameters()$saem
predictions <- computePredictions( individualParameters = individualParamsForAllIndiv[ids,],
individualIds = ids )
obsInfo = getObservationInformation()
allIds = unique(obsInfo$y$id)
obsPred = cbind(obsInfo$y[obsInfo$y$id %in% allIds[ids]], y_pred = predictions$Cc)