rdbinselect
employs the results in Calonico, Cattaneo and Titiunik (2014d) to implement several novel data-driven optimal choices of evenly-spaced and quantile-based bins used to approximate the underlying regression functions by local sample means, which is then used to construct the RD plots usually found in empirical applications.
Calonico, Cattaneo and Titiunik (2014b) provides an introduction to this command. For a review on RD methods see Imbens and Lemieux (2008), Lee and Lemieux (2010), Dinardo and Lee (2011), Calonico, Cattaneo and Titiunik (2014a), and references therein.
A companion Stata
package is described in Calonico, Cattaneo and Titiunik (2014c).rdbinselect(y, x, data, subset = NULL, c = 0, p = 4,
numbinl = NULL, numbinr = NULL, binselect = "es",
lowerend = NULL, upperend = NULL, scale = 1, hide = FALSE,
par=NULL, title = NULL, x.label = NULL, y.label = NULL,
x.lim = NULL, y.lim = NULL, model = FALSE, frame = FALSE)
rdbwselect
, rdrobust
x<-runif(1000,-1,1)
y<-5+3*x+2*(x>=0)+rnorm(1000)
rdbinselect(y,x)
Run the code above in your browser using DataLab