psw.wt {PSW}

Propensity score weighting estimator

Description

psw.wt is used to estimate the weighted treatment effect estimator (without double robustness).

Usage

psw.wt(data, form.ps, weight, out.var, family = "gaussian", K = 4)

Arguments

data

data frame to be used.

form.ps

propensity score model.

weight

weighting method to be used. Available methods are "ATE", "ATT", "ATC", "MW", "OVERLAP", and "TRAPEZOIDAL".

out.var

outcome variable.

family

outcome family, either "gaussian" or "binomial". family="gaussian" by default.

K

value of K in \omega(e_i) = min(1, K min(e_i, 1-e_i)) for "TRAPEZOIDAL" weight. The estimand is closer to the average treatment effect (ATE) with larger value of K. K=4 by default.

Details

psw.wt is used to estimate the weighted estimator, \hat{\Delta}, and make inference using the sandwich variance estimator that takes into account the sampling variability in the estimated propensity score.

Value

A list of weighting method, fitted propensity score model, estimated propenstity scores, estimated propensity score weights, weighted estimator and standard error estimator

weight

weighting method.

ps.model

object returned by fitting the propensity score model using glm with "binomial" family.

ps.hat

estimated propensity score.

W

estimated propensity score weight.

est.wt

weighted estimator for mean difference when family = "gaussian".

std.wt

standard error for est.wt.

est.risk.wt

weighted estimator for risk difference when family = "binomial".

std.risk.wt

standard error for est.risk.wt.

est.rr.wt

weighted estimator for relative risk when family = "binomial".

std.rr.wt

standard error for est.rr.wt.

est.or.wt

weighted estimator for odds ratio when family = "binomial".

std.or.wt

standard error for est.or.wt.

est.lor.wt

weighted estimator for log odds ratio when family = "binomial".

std.lor.wt

standard error for est.lor.wt.

See Also

psw

Examples

# Load the test data set
data(test_data);
# Propensity score model
form.ps <- "Z ~ X1 + X2 + X3 + X4";
tmp <- psw.wt( data = test_data, weight = "ATE", form.ps = form.ps,
out.var = "Y", family = "gaussian" );