partial_dependence_probit: Partial Dependence Function for SoftBART Probit Regression

In SoftBart: Implements the SoftBart Algorithm

Partial Dependence Function for SoftBART Probit Regression

Description

Computes the partial dependence function for a given covariate at a given set of covariate values for the probit model.

Usage

partial_dependence_probit(fit, test_data, var_str, grid)

Arguments

fit	A fitted model of type softbart_probit.
test_data	A data set used to form the baseline distribution of covariates for the partial dependence function.
var_str	A string giving the variable name of the predictor to compute the partial dependence function for.
grid	The values of the predictor to compute the partial dependence function at.

Value

Returns a list with the following components:

• pred_df: a data frame containing columns for a MCMC iteration ID (sample), the value on the grid, and the partial dependence function value.

• p: a matrix containing the same information as pred_df, with the rows corresponding to iterations and columns corresponding to grid values.

• grid: the grid used as input.

SoftBart documentation built on June 8, 2025, 9:40 p.m.