bartMachine (version 1.2.6)

get_var_counts_over_chain: Get the Variable Inclusion Counts

Description

Computes the variable inclusion counts for a BART model.

Usage

get_var_counts_over_chain(bart_machine, type = "splits")

Value

Returns a matrix of counts of each predictor across all trees by Gibbs sample. Thus, the dimension is num_interations_after_burn_in

by p (where p is the number of predictors after dummifying factors and adding missingness dummies if specified by use_missing_data_dummies_as_covars).

Arguments

bart_machine

An object of class ``bartMachine''.

type

If ``splits'', then the number of times each variable is chosen for a splitting rule is computed. If ``trees'', then the number of times each variable appears in a tree is computed.

Author

Adam Kapelner and Justin Bleich

See Also

get_var_props_over_chain

Examples

if (FALSE) {

#generate Friedman data
set.seed(11)
n  = 200 
p = 10
X = data.frame(matrix(runif(n * p), ncol = p))
y = 10 * sin(pi* X[ ,1] * X[,2]) +20 * (X[,3] -.5)^2 + 10 * X[ ,4] + 5 * X[,5] + rnorm(n)

##build BART regression model
bart_machine = bartMachine(X, y, num_trees = 20)

#get variable inclusion counts
var_counts = get_var_counts_over_chain(bart_machine)
print(var_counts)
}

Run the code above in your browser using DataCamp Workspace