partialDependence: Compute partial dependence

In brandmaier/semtree: Recursive Partitioning for Structural Equation Models

Compute partial dependence

Description

Compute the partial dependence of a predictor, or set of predictors, on a model parameter.

Usage

partialDependence(
  x,
  data,
  reference.var,
  support = 20,
  points = NULL,
  mc = NULL,
  FUN = "median",
  ...
)

Arguments

x	An object for which a method exists
data	Optional data.frame that was used to train the model.
reference.var	Character vector, referring to the (independent) reference variable or variables for which partial dependence is calculated. Providing two (or more) variables allows for probing interactions, but note that this is computationally expensive.
support	Integer. Number of grid points for interpolating the reference.var. Alternatively, use points for one or more variables named in reference.var.
points	Named list, with elements corresponding to reference.var . Use this argument to provide specific points for which to obtain marginal dependence values; for example, the mean and +/- 1SD of reference.var.
mc	Integer. If mc is not NULL, the function will sample mc number of rows from data with replacement, to estimate marginal dependency using Monte Carlo integration. This is less computationally expensive.
FUN	Character string with function used to integrate predictions across all elements of x.
...	Extra arguments passed to FUN.

Author(s)

Caspar J. Van Lissa, , Andreas M. Brandmaier

brandmaier/semtree documentation built on April 18, 2024, 3:24 a.m.