The nprobust package provides Stata and R implementations of bandwidth selection, point estimation and inference procedures for nonparametric kernel-based density and local polynomial methods.
This work is supported by the National Science Foundation through grant SES-1459931.
For technical, methodological and implementation details see the following papers (and references therein):
- Calonico, Cattaneo and Farrell (2018): On the Effect of Bias Estimation on Coverage Accuracy in Nonparametric Inference, Journal of the American Statistical Association 113(522): 767-779. [Supplemental Appendix]
- Calonico, Cattaneo and Farrell (2018): Coverage Error Optimal Confidence Intervals. [Supplemental Appendix]
- Calonico, Cattaneo and Farrell (2018): nprobust: Nonparametric Kernel-Based Estimation and Robust Bias-Corrected Inference, Journal of Statistical Software, forthcoming.
Implementation in Stata:
- To install/update in Stata type:
· net install nprobust, from(https://sites.google.com/site/nppackages/nprobust/stata) replace
or
· github install iphone7725/nprobust
- Help files: kdrobust, kdbwselect, lprobust, lpbwselect -- Replication files: do-file, nprobust_data
- Repository for manual installation: https://sites.google.com/site/nppackages/nprobust/stata
Implementation in R:
- To install/update in R type:
· install.packages('nprobust')
- Manual -- Replication files: R-script, nprobust_data
- CRAN repository
Last update: May 1, 2018.