Hi,
I have a question about implementing a regression discontinuity design (RDD) where the running variable is age (discrete). Specifically, I am looking for a Stata package or command that can help determine the optimal bandwidth in this context.
I have reviewed the CCT approach, but it seems to focus on continuous running variables. I also looked into "Optimized Regression Discontinuity Designs" by Imbens and Wager (REStat 2019), and "Inference in Regression Discontinuity Designs with a Discrete Running Variable" by Kolesár and Rothe (AER 2018). However, it appears that the rdbwselect package doesn’t incorporate these methods or is primarily designed for continuous running variables.
In studies that use discrete age as the running variable, bandwidths are typically pre-determined rather than optimized. If anyone has worked on similar topics or knows of an appropriate method or package for selecting an optimal bandwidth with a discrete running variable, I would greatly appreciate your suggestions.
Thank you!
I have a question about implementing a regression discontinuity design (RDD) where the running variable is age (discrete). Specifically, I am looking for a Stata package or command that can help determine the optimal bandwidth in this context.
I have reviewed the CCT approach, but it seems to focus on continuous running variables. I also looked into "Optimized Regression Discontinuity Designs" by Imbens and Wager (REStat 2019), and "Inference in Regression Discontinuity Designs with a Discrete Running Variable" by Kolesár and Rothe (AER 2018). However, it appears that the rdbwselect package doesn’t incorporate these methods or is primarily designed for continuous running variables.
In studies that use discrete age as the running variable, bandwidths are typically pre-determined rather than optimized. If anyone has worked on similar topics or knows of an appropriate method or package for selecting an optimal bandwidth with a discrete running variable, I would greatly appreciate your suggestions.
Thank you!
