Regression discontinuity command "rdrobust" with survey weights

[Santiago Valdivieso]

I am using rdrobust command in a household survey. As the survey uses a complex sampling design, I need to use consider survey weights in the estimation. Is there a way to do this in Stata? .

I see that there is a "weights" option in the command, but I would guess it refers to the kernel weights rather than survey weights.

A similar question was posted years ago, but no one answered it

[Maxime Bercholz]

Hi Santiago,

I don't have an answer to your question (other than to say that the unit weights specified in rdrobust's weights() option multiply the kernel weights), but did you find a way to run rdrobust (or estimate local kernel regressions) accounting for your complex sampling design?