marginal_effects: Marginal Effects for Modified Ordered Random Forests

In morf: Modified Ordered Random Forest

Marginal Effects for Modified Ordered Random Forests

Description

Nonparametric estimation of marginal effects using an morf object.

Usage

marginal_effects(
  object,
  data = NULL,
  eval = "atmean",
  bandwitdh = 0.01,
  inference = FALSE
)

Arguments

object	An morf object.
data	Data set of class data.frame to estimate marginal effects. It must contain at least the same covariates used to train the forests. If NULL, marginal effects are estimated on object$full_data.
eval	Evaluation point for marginal effects. Either "mean", "atmean" or "atmedian".
bandwitdh	How many standard deviations x_up and x_down differ from x.
inference	Whether to extract weights and compute standard errors. The weights extraction considerably slows down the program.

Details

marginal_effects can estimate mean marginal effects, marginal effects at the mean, or marginal effects at the median, according to the eval argument.

The routine assumes that covariates with more than ten unique values are continuous. Otherwise, covariates are assumed to be discrete.

Value

Object of class morf.marginal.

Author(s)

Riccardo Di Francesco

See Also

morf

Examples

## Load data from orf package.
set.seed(1986)

library(orf)
data(odata)
odata <- odata[1:200, ] # Subset to reduce elapsed time.

y <- as.numeric(odata[, 1])
X <- as.matrix(odata[, -1])

## Fit morf . Use large number of trees.
forests <- morf(y, X, n.trees = 4000)

## Marginal effects at the mean.
me <- marginal_effects(forests, eval = "atmean")
print(me)
summary(me)

## LATEX.
print(me, latex = TRUE)


## Compute standard errors. This requires honest forests.
honest_forests <- morf(y, X, n.trees = 4000, honesty = TRUE)
honest_me <- marginal_effects(honest_forests, eval = "atmean", inference = TRUE)
honest_me$standard.errors
honest_me$p.values # These are not corrected for multiple hypotheses testing!

print(honest_me, latex = TRUE)

morf documentation built on March 31, 2023, 8:14 p.m.