Regression Discontinuity Design Application

Estimation of both single- and multiple-assignment Regression Discontinuity Designs (RDDs). Provides both parametric (global) and non-parametric (local) estimation choices for both sharp and fuzzy designs, along with power analysis and assumption checks. Introductions to the underlying logic and analysis of RDDs are in Thistlethwaite, D. L., Campbell, D. T. (1960) and Lee, D. S., Lemieux, T. (2010) .


Travis-CI Build Status Downloads Cran Build

Overview

rddapp provides a set of functions for the analysis of the regression-discontinuity design (RDD).

The three main parts are:

  • estimation of effects of interest
  • power analysis
  • assumption checks

Estimation

The package estimates treatment effects from RDDs, for the following designs and approaches:

  • parametric RDD with single assignment variables, both sharp and fuzzy designs
  • non-parametric RDD with single assignment variables, both sharp and fuzzy designs
  • parametric RDDs with two assignment variables, both sharp and fuzzy designs, using univariate, centering, and frontier approaches

Power analysis

Given input from the user about desired Type I error rate, and assumptions about the population, the package allows estimation of power for the following designs:

  • single assignment RDDs (both sharp and fuzzy) using both parametric and non-parametric estimation
  • multiple-assignment RDDs (both sharp and fuzzy) using various parametric models

Assumption checks

The package allows the user to perform a variety of assumption and sensitivity checks.

  • McCrary's sorting test on the assignment variable
  • Sensitivity to the chosen bandwidth in non-parametric estimation
  • Placebo tests to examine treatment effects at values away from the cut-off
  • Discontinuities in the treatment probability at cut-off
  • Discontinuities for baseline covariates

Installation

install.packages("rddapp")
 
# Or the development version from GitHub:
# install.packages("devtools")
devtools::install_github("felixthoemmes/rddapp")

News

rddapp 0.1.0

  • Initial commit.

  • Add R functions: estimation, power analysis, and assumption checks.

  • Add R tests: estimation and assumption checks.

  • Add CARE dataset.

rddapp 1.0.0

  • Add shiny app.

  • Add R function: shiny app launch.

rddapp 1.1.0

  • Shiny app has new download buttons, screenshots, and a draft manual

  • Several bugfixes in functions that used the bootstrap (bad integrand issue fixed)

  • Summary functions have been updated to displays results in more consistent format

  • Summary functions are now including confidence intervals by default

  • Stability fixes for dc_est function

  • Default of t.design was changed to NULL to force user to always specify design

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("rddapp")

1.1.0 by Felix Thoemmes, a year ago


Browse source code at https://github.com/cran/rddapp


Authors: Ze Jin [aut] , Wang Liao [aut] , Irena Papst [aut] , Wenyu Zhang [aut] , Felix Thoemmes [aut, cre] , Drew Dimmery [ctb] (Implementation of functions in R package rdd) , Bastiaan Quast [ctb] (Implementation of functions in R package rddtools) , Matthieu Stigler [ctb] (Implementation of functions in R package rddtools)


Documentation:   PDF Manual  


Task views: Econometrics


GPL (>= 2) license


Imports AER, sandwich, lmtest, Formula, shiny, R.utils

Suggests rdd, rddtools, foreign, devtools, testthat, roxygen2, knitr, rmarkdown, DT


See at CRAN