Setting Options for tmlenet

Share:

Description

Additional options that control the estimation algorithm in tmlenet package

Usage

1
2
3
tmlenet_options(useglm = FALSE, parfit = FALSE,
  bin.method = c("equal.len", "equal.mass", "dhist"), nbins = NA,
  maxncats = 20, poolContinVar = FALSE, maxNperBin = 1000)

Arguments

useglm

Set to FALSE to estimate with speedglm.wfit and TRUE for glm.fit.

parfit

Default is FALSE. Set to TRUE to use foreach package and its functions foreach and dopar to perform parallel logistic regression fits and predictions for discretized continuous outcomes. This functionality requires registering a parallel backend prior to running tmlenet function, e.g., using doParallel R package and running registerDoParallel(cores = ncores) for integer ncores parallel jobs. For an example, see a test in "./tests/RUnit/RUnit_tests_04_netcont_sA_tests.R".

bin.method

The method for choosing bins when discretizing and fitting the conditional continuous summary exposure variable sA. The default method is "equal.len", which partitions the range of sA into equal length nbins intervals. Method "equal.mass" results in a data-adaptive selection of the bins based on equal mass (equal number of observations), i.e., each bin is defined so that it contains an approximately the same number of observations across all bins. The maximum number of observations in each bin is controlled by parameter maxNperBin. Method "dhist" uses a mix of the above two approaches, see Denby and Mallows "Variations on the Histogram" (2009) for more detail.

nbins

Set the default number of bins when discretizing a continous outcome variable under setting bin.method = "equal.len". If left as NA the total number of equal intervals (bins) is determined by the nearest integer of nobs/maxNperBin, where nobs is the total number of observations in the input data.

maxncats

Max number of unique categories a categorical variable sA[j] can have. If sA[j] has more it is automatically considered continuous.

poolContinVar

Set to TRUE for fitting a pooled regression which pools bin indicators across all bins. When fitting a model for binirized continuous outcome, set to TRUE for pooling bin indicators across several bins into one outcome regression?

maxNperBin

Max number of observations per 1 bin for a continuous outcome (applies directly when bin.method="equal.mass" and indirectly when bin.method="equal.len", but nbins = NA).

Value

Invisibly returns a list with old option settings.

See Also

print_tmlenet_opts

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.