perm_test: Permutation test Test the null hypothesis P(X|I=0) =...

In optinterv: Optimal Intervention

Description

Permutation test Test the null hypothesis P(X|I=0) = P(X|I=1), using permutation test.

Usage

perm_test(estimates, wgt, wgt1, X, n.quant, n.perm = 1000, Y = NULL,
  control = NULL, func = "non_parm")

Arguments

estimates	point estimates of the percentile distance between P(X|I=0) & P(X|I=1).
wgt	an optional vector of weights.
wgt1	weights under I = 1.
X	numeric data frame or matrix of factors to be considered.
n.quant	number of quantiles to use when calculating CDF distance.
n.perm	number of permutations to permute from wgt1.
Y	outcome vector (must be numeric without NA's).
control	numeric data frame or matrix of factors to control for. these are factors that we can't consider while looking for the optimal intervention (e.g. race).
func	either "non_parm" or "nn". for "nn", weights are recalculated for each permutation, and thus Y and control are needed. the default is "non_parm", and Y and control aren't needed.

Value

vector of p values.

optinterv documentation built on March 26, 2020, 7:05 p.m.