twofold_test.pk {invivoPKfit} | R Documentation |
Evaluate whether data and predictions are within two-fold of mean or concentration, respectively
Description
At each timepoint across CvT experimental data, there are three ways that data may be presented. These can be found as either: - multiple individual observations - single individual observation - summarized group of observations (mean concentration and standard deviation)
Usage
## S3 method for class 'pk'
twofold_test(
obj,
sub_pLOQ = TRUE,
suppress.messages = NULL,
model = NULL,
method = NULL,
...
)
Arguments
obj |
A pk object. |
sub_pLOQ |
TRUE (default): Substitute all predictions below the LOQ with the LOQ before computing fold errors. FALSE: do not. Only used if 'obj' has been fitted and predictions are possible. |
suppress.messages |
Logical: whether to suppress message printing. If NULL (default), uses the setting in 'obj$settings_preprocess$suppress.messages'. |
model |
Optional: Specify one or more of the fitted models for which to make predictions. If NULL (the default), predictions will be returned for all of the models in 'object$stat_model'. |
method |
Optional: Specify one or more of the [optimx::optimx()] methods for which to make predictions. If NULL (the default), predictions will be returned for all of the models in 'object$settings_optimx$method'. |
... |
Additional arguments. Currently unused. |
Details
For the purposes of this calculations we largely divide the data into two groups, those with individual observations, where N_Subjects == 1, and the summarized group of observations.
First this creates mean-normalized concentrations for individual data. Then it summarizes data (individual & summarized) by 'mean' and 'sd'. It tests whether predictions are within two-fold of mean, in the latter case whether the 95
Furthermore if 'pk' object 'status == 5' then it calculates the model error by evaluating _prediction/concentration_ at each timepoint for all data. Each test is done for data from individual subject observations and for all data by summarizing the observations.
Only non-excluded detects are included in this analysis.
Value
A list of data frames.
Author(s)
Gilberto Padilla Mercado