Cutoff Sensitivity Simulation for Regression Discontinuity
rd_sens_cutoff.Rdrd_sens_cutoff refits the supplied model with varying cutoff(s).
All other aspects of the model, such as the automatically calculated bandwidth, are held constant.
Arguments
- object
An object returned by
rd_estorrd_impute.- cutoffs
A numeric vector of cutoff values to be used for refitting an
rdobject.
Value
rd_sens_cutoff returns a dataframe containing the estimate est and standard error se
for each cutoff value (A1). Column A1 contains varying cutoffs
on the assignment variable. The model column contains the parametric model (linear, quadratic, or cubic) or
non-parametric bandwidth setting (Imbens-Kalyanaraman 2012 optimal, half, or double) used for estimation.
References
Imbens, G., Kalyanaraman, K. (2012). Optimal bandwidth choice for the regression discontinuity estimator. The Review of Economic Studies, 79(3), 933-959. https://academic.oup.com/restud/article/79/3/933/1533189.
Examples
set.seed(12345)
x <- runif(1000, -1, 1)
cov <- rnorm(1000)
y <- 3 + 2 * x + 3 * cov + 10 * (x >= 0) + rnorm(1000)
rd <- rd_est(y ~ x | cov, t.design = "geq")
rd_sens_cutoff(rd, seq(-.5, .5, length.out = 10))
#> est se A1 model
#> 1 0.07969423 0.2444090 -0.50000000 linear
#> 2 -3.56425289 0.3484931 -0.50000000 quadratic
#> 3 -2.14631083 0.4574047 -0.50000000 cubic
#> 4 -2.25549684 0.2466853 -0.50000000 optimal
#> 5 0.08617364 0.2389141 -0.50000000 half
#> 6 -1.05563755 0.2115650 -0.50000000 double
#> 7 1.13036935 0.2641814 -0.38888889 linear
#> 8 -3.01554092 0.2816176 -0.38888889 quadratic
#> 9 -3.15717647 0.3541280 -0.38888889 cubic
#> 10 -2.29963553 0.2109532 -0.38888889 optimal
#> 11 -0.44725446 0.2049669 -0.38888889 half
#> 12 -0.01664455 0.2244850 -0.38888889 double
#> 13 3.14369677 0.3310038 -0.27777778 linear
#> 14 -1.66674376 0.3016680 -0.27777778 quadratic
#> 15 -3.91768354 0.4345337 -0.27777778 cubic
#> 16 -1.14834321 0.2357961 -0.27777778 optimal
#> 17 -2.00224706 0.3039055 -0.27777778 half
#> 18 2.03843241 0.3113976 -0.27777778 double
#> 19 5.23948286 0.3804960 -0.16666667 linear
#> 20 0.87627157 0.4205371 -0.16666667 quadratic
#> 21 -3.06449227 0.3957340 -0.16666667 cubic
#> 22 1.68951550 0.3927183 -0.16666667 optimal
#> 23 -2.46201237 0.3017299 -0.16666667 half
#> 24 4.38509932 0.3993719 -0.16666667 double
#> 25 8.60263592 0.3126928 -0.05555556 linear
#> 26 6.95763696 0.5682479 -0.05555556 quadratic
#> 27 4.64096453 0.7448366 -0.05555556 cubic
#> 28 7.43246057 0.4904542 -0.05555556 optimal
#> 29 5.05961343 0.7054854 -0.05555556 half
#> 30 8.32612885 0.3574042 -0.05555556 double
#> 31 7.54145063 0.3672640 0.05555556 linear
#> 32 5.06056419 0.5491753 0.05555556 quadratic
#> 33 2.26934794 0.5770854 0.05555556 cubic
#> 34 5.69759456 0.5048512 0.05555556 optimal
#> 35 2.76292037 0.5476351 0.05555556 half
#> 36 7.09121133 0.4075403 0.05555556 double
#> 37 4.73711356 0.3838563 0.16666667 linear
#> 38 0.15873621 0.3938900 0.16666667 quadratic
#> 39 -3.58197805 0.4361977 0.16666667 cubic
#> 40 0.94210596 0.3604900 0.16666667 optimal
#> 41 -2.74438672 0.3349725 0.16666667 half
#> 42 3.83493989 0.3930972 0.16666667 double
#> 43 2.44021253 0.3070641 0.27777778 linear
#> 44 -2.01011054 0.2888192 0.27777778 quadratic
#> 45 -3.33888035 0.4000231 0.27777778 cubic
#> 46 -1.47547195 0.2187244 0.27777778 optimal
#> 47 -1.35816810 0.2691661 0.27777778 half
#> 48 1.35284668 0.2765852 0.27777778 double
#> 49 0.74623053 0.2460348 0.38888889 linear
#> 50 -3.16706426 0.2664022 0.38888889 quadratic
#> 51 -2.74805736 0.3310677 0.38888889 cubic
#> 52 -2.38293452 0.1992077 0.38888889 optimal
#> 53 -0.27524311 0.1742558 0.38888889 half
#> 54 -0.36324795 0.2028941 0.38888889 double
#> 55 -0.32689908 0.2209892 0.50000000 linear
#> 56 -3.41397231 0.2968565 0.50000000 quadratic
#> 57 -1.39376696 0.3504643 0.50000000 cubic
#> 58 -2.05049172 0.2155820 0.50000000 optimal
#> 59 -0.02753801 0.1967451 0.50000000 half
#> 60 -1.34703640 0.1892384 0.50000000 double
#> 61 10.04882040 0.1190290 0.00000000 linear
#> 62 10.06839238 0.1780133 0.00000000 quadratic
#> 63 9.82936267 0.2253520 0.00000000 cubic
#> 64 10.04667463 0.1413530 0.00000000 optimal
#> 65 9.89572808 0.1885718 0.00000000 half
#> 66 10.05245913 0.1199932 0.00000000 double