Robust Local Polynomial Methods for Heterogenous Treatment Effects in RD Designs

The package rdhte implements estimation, inference, and bandwidth selection procedures for heterogeneous treatment effects in Regression Discontinuity (RD) designs using local polynomial methods.

• rdhte: point estimation and robust bias-corrected inference for conditional RD treatment effects.
• rdbwhte: data-driven bandwidth selection for RD heterogeneous treatment effect estimation.
• rdhte_lincom: post-estimation tests for linear combinations of heterogeneous treatment effect parameters.

Python Implementation

To install/update in Python type:

pip install rdhte

• Help: PYPI repository.

• Examples/data: rdhte illustration, rdhte data.

R Implementation

To install/update in R type:

install.packages('rdhte')

• Help: R Manual, CRAN repository.

• Examples/data: rdhte illustration, rdhte data.

Stata Implementation

To install/update in Stata type:

net install rdhte, from(https://raw.githubusercontent.com/rdpackages/rdhte/main/stata) replace

• Help: rdhte, rdbwhte, rdhte_lincom, rdhte_plot.

• Replication: rdhte illustration, rdhte data.

References

For overviews and introductions, see the rdpackages website.

Software and Implementation

• Calonico, Cattaneo, Farrell, Palomba and Titiunik (2026): rdhte: Conditional Average Treatment Effects in RD Designs. Working paper.

Technical and Methodological

• Calonico, Cattaneo, Farrell, Palomba and Titiunik (2026): Treatment Effect Heterogeneity in Regression Discontinuity Designs. Working paper. Supplemental Appendix.

Dataset for Replication

• Granzier, Pons and Tricaud (2023): Coordination and Bandwagon Effects: How Past Rankings Shape the Behavior of Voters and Candidates, American Economic Journal: Applied Economics 15(4): 177-217.

Funding

This work was supported in part by the National Science Foundation through grants SES-2019432 and SES-2241575.