rddensity package provides Stata and R implementations of manipulation tests employing local polynomial density estimation methods. This method is useful for falsification of Regression Discontinuity Designs, as well as for testing for self-selection or sorting in other contexts. This implementation provides hypothesis tests and bandwidth selectors for manipulation testing.
This package was first released in Spring 2017, and had one major upgrade in Summer 2020.
lpdensity; (iii) mass points in running variable adjustments; (iv) bandwidth selection adjustments for too few mass points in and/or overshooting of the support of the running variable; (v) density discontinuity plots with histogram and/or confidence bands; and (vi) binomial testing near cutoff as complementary discontinuity testing following results in
rdlocrandmethods (see references there for details).
To install/update in Stata type:
net install rddensity, from(https://raw.githubusercontent.com/rdpackages/rddensity/master/stata) replace
lpdensity, to install/update in Stata type:
net install lpdensity, from(https://raw.githubusercontent.com/nppackages/lpdensity/master/stata) replace
To install/update in R type:
lpdensity, to install/update in R type:
For source code and related files, visit
For overviews and introductions, see rdpackages website.
Cattaneo, Jansson and Ma (2018): Manipulation Testing based on Density Discontinuity.
Stata Journal 18(1): 234-261.
Cattaneo, Jansson and Ma (2021): lpdensity: Local Polynomial Density Estimation and Inference.
Journal of Statistical Software, forthcoming.