| Title: | Diagnostic Tool by Multiple Imputation for Regression Discontinuity Designs |
| Version: | 0.1.0 |
| Description: | Estimates local average treatment effects based on regression discontinuity designs (RDD) and multiple imputation regression discontinuity designs (MIRDD). It provides diagnostic tools for RDD by comparing results with those from MIRDD, as proposed in Takahashi (2023) <doi:10.1080/03610918.2021.1960374>. |
| License: | GPL-3 |
| Encoding: | UTF-8 |
| Imports: | Amelia, rdrobust, stats, graphics |
| RoxygenNote: | 7.3.3 |
| NeedsCompilation: | no |
| Packaged: | 2026-03-14 08:52:40 UTC; masay |
| Author: | Masayoshi Takahashi [aut, cre] |
| Maintainer: | Masayoshi Takahashi <mtakahashi615@g.chuo-u.ac.jp> |
| Repository: | CRAN |
| Date/Publication: | 2026-03-19 14:10:08 UTC |
Diagnostic Tool by Multiple Imputation for Regression Discontinuity Designs
Description
Estimates local average treatment effects based on regression discontinuity designs (RDD) and multiple imputation regression discontinuity designs (MIRDD). It provides diagnostic tools for RDD by comparing results with those from MIRDD.
Usage
MIdiagRDD(
y,
x,
cut,
seed = 1,
M1 = 100,
M2 = 5,
M3 = 1,
p2s1 = 1,
emp = 0,
bw = "mserd",
ker = "triangular",
bwidth = 1,
p1 = 1,
conf = 95,
upper = 1,
covs1 = NULL,
up = NULL,
lo = NULL
)
Arguments
y |
A numeric vector of the outcome variable. |
x |
A numeric vector of the assignment variable (forcing variable). |
cut |
A numeric value indicating the cutoff point. |
seed |
A seed number for reproducibility. Default is 1. |
M1 |
Number of imputations for MIRDD. Default is 100. |
M2 |
Number of imputations for visualization. Default is 5. |
M3 |
Number of imputations for specific plots. Default is 1. |
p2s1 |
Integer for Amelia's p2s argument (0, 1, or 2). Default is 1. |
emp |
Amelia's empriical prior. Default is 0. |
bw |
Bandwidth selection method for rdrobust. Default is "mserd". |
ker |
Kernel function for rdrobust. Default is "triangular". |
bwidth |
Scaling factor for bandwidth. Default is 1. |
p1 |
Polynomial order for rdrobust. Default is 1. |
conf |
Confidence level (0-100). Default is 95. |
upper |
If 1 (default), treatment is x >= cut. If 0, treatment is x < cut. |
covs1 |
Optional covariates for Amelia. |
up |
Optional upper bound for plots. |
lo |
Optional lower bound for plots. |
Value
No return value, called for side effects (plotting and printing results).
Examples
# Example usage with dummy data
x <- runif(100, -1, 1)
y <- 0.5 * x + (x >= 0) + rnorm(100, 0, 0.1)
MIdiagRDD(y = y, x = x, cut = 0)