# Simple example of regression discontinuity design
set.seed(42)
n = 1000; threshold = 0
X = runif(n, -1, 1)
W = as.numeric(X >= threshold)
Y = (1 + 2*W)*(1 + X^2) + 1 / (1 + exp(X)) + rnorm(n, sd = .5)
out = plrd(Y, X, threshold)
print(out)
#> Partially linear regression discontinuity inference:
#> Threshold: 0
#> Lipschitz constant: 7.302
#> Max bias: 0.05186
#> Sampling SE: 0.0856
#> Effective sample size: 990
#> Confidence level: 95%
#>
#> Estimate CI Lower CI Upper p-value
#> RD Estimate 2.022 1.828 2.217 1.461e-117
# default plot: scatterplot of original data with black curves showing
# representative regression functions in our data-driven function class
plot(out)
# plot of plrd weights (two sets of weights because we use cross-fitting)
plot(out, type = "weights")
# fancy plot option that combines the previous two plots
plot(out, type = "combined")
