latrendCV: Cluster longitudinal data over k folds

View source: R/latrend.R

latrendCVR Documentation

Cluster longitudinal data over k folds

Description

Apply k-fold cross validation for internal cluster validation. Creates k random subsets ("folds") from the data, estimating a model for each of the k-1 combined folds.

Usage

latrendCV(
  method,
  data,
  folds = 10,
  seed = NULL,
  parallel = FALSE,
  errorHandling = "stop",
  envir = NULL,
  verbose = getOption("latrend.verbose")
)

Arguments

method

An lcMethod object specifying the longitudinal cluster method to apply, or the name (as character) of the lcMethod subclass to instantiate.

data

A data.frame.

folds

The number of folds. Ten folds by default.

seed

The seed to use. Optional.

parallel

Whether to enable parallel evaluation. See latrend-parallel. Method evaluation and dataset transformation is done on the calling thread.

errorHandling

Whether to "stop" on an error, or to ⁠"remove'⁠ evaluations that raised an error.

envir

The environment in which to evaluate the method arguments via compose(). If the data argument is of type call then this environment is also used to evaluate the data argument.

verbose

The level of verbosity. Either an object of class Verbose (see R.utils::Verbose for details), a logical indicating whether to show basic computation information, a numeric indicating the verbosity level (see Verbose), or one of c('info', 'fine', 'finest').

Value

A lcModels object of containing the folds training models.

See Also

Other longitudinal cluster fit functions: latrendBatch(), latrendBoot(), latrendRep(), latrend()

Other validation methods: createTestDataFolds(), createTestDataFold(), createTrainDataFolds(), latrendBoot(), lcModel-data-filters

Examples

data(latrendData)
method <- lcMethodLMKM(Y ~ Time, id = "Id", time = "Time")

if (require("caret")) {
  model <- latrendCV(method, latrendData, folds = 5, seed = 1)

  model <- latrendCV(method, subset(latrendData, Time < .5), folds = 5)
}

latrend documentation built on March 31, 2023, 5:45 p.m.