In frankpopham/estimateit: Estimate effects for binary exposures and outcomes from a Weightit object

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

estimateit

The goal of estimateit is to estimate marginal effects using inverse probability weights generated by weightit. Effects are generated using svyglm and svycontrast from thesurvey package. They include the effect for the treated, the effect for the untreated or control, their difference, their relative difference, and their odds ratio. As the previous sentence implies estimateit only works with binary exposures and binary outcomes at present.

Installation

You can install the development version of estimateit from GitHub with:

# install.packages("devtools")
devtools::install_github("frankpopham/estimateit")

Examples

## A simple example
library(WeightIt)
library(tibble)
library(tidyr)
dfvi <- tibble(
 C = rep(0:1, each = 4),
 X = rep(0:1, times = 4),
 Y = rep(0:1, times = 2, each = 2),
 N = c(96, 36, 64, 54, 120, 120, 30, 480)
 ) %>%
 uncount(N)
 W1 <- weightit(X ~ C, data = dfvi,
               method = "ps", estimand = "ATE")
summary(W1)
E1 <- estimateit(weightitobj=W1, outcome=Y, data=dfvi)
E1

frankpopham/estimateit documentation built on Jan. 30, 2023, 12:52 a.m.