Implement a useful bootstrap feature - Githubissues

Calvagone / campsis

A generic PK/PD simulation platform based on rxode2 and mrgsolve engines.

https://calvagone.github.io/

GNU General Public License v3.0

8 stars 3 forks source link

Implement a useful bootstrap feature #76

Closed luyckxn closed 2 years ago

luyckxn commented 2 years ago

Current distribution 'BootstrapDistribution' has a couple of limitations:

Because it must be used within a single covariate, it cannot deal with multiple covariates of an individual (unless using random=FALSE and replacement=TRUE)
In that latter case (unless random=FALSE and replacement=TRUE), there is no randomness, and the exact same covariates are used for all replicates, which is not what we want (in general)

We need an object Bootstrap that takes a data frame of covariates and that can be added to the dataset. It must have the arguments:

random: draw the subjects' covariates at random (use a different seed for each replicate)
replacement: if subjects can be drawn with replacement