Package: rd2dnet 0.1.0

rd2dnet: Boundary Discontinuity Designs on Networks

Estimation and inference for boundary discontinuity designs on street networks. Extends the location-based framework of 'rd2d' (Cattaneo, Titiunik and Yu, <doi:10.48550/arXiv.2505.07989>) from Euclidean coordinates to network-native coordinates (signed network distance to the boundary and offset along it) and user-defined network metrics (edge costs, travel time, road-class transition penalties, precomputed matrices). Provides side-specific local polynomial estimates of boundary average treatment effect curves with robust bias-corrected inference (Calonico, Cattaneo and Titiunik, 2014, <doi:10.3982/ECTA11757>), uniform confidence bands, weighted and largest aggregate effects, fuzzy designs, covariates, derivative targets, heteroskedasticity-robust, cluster and network Conley (1999, <doi:10.1016/S0304-4076(98)00084-0>) standard errors, a port of the 'rd2d' direct plug-in bandwidth selector, and a network-normalised version of the local polynomial density test of Cattaneo, Jansson and Ma (2020, <doi:10.1080/01621459.2019.1635480>).

Authors:Zach Winship [aut, cre, cph], Matias D. Cattaneo [ctb, cph], Rocío Titiunik [ctb, cph], Ruiqi Rae Yu [ctb, cph], Michael Jansson [ctb, cph], Xinwei Ma [ctb, cph]