0

I am looking for the formula of a regression discontinuity, where the independent variable is the dummy (0, 1) and the dependent variable is continuous. Furthermore, I would like to add several confounders to this formula.

Many books point me to this formula, which is not sufficient for my study:

Υ = ∝ + Dτ + Xβ + ε

where
D = dummy
τ = treatment
ε = error

Could someone point me into the right direction? I am really struggling with adding the dummy as independent variable, and including the confounders in the formula.

Your help is very much appreciated!

  • I'm voting to close this question as off-topic because it is not about programming. – High Performance Mark Jan 18 '18 at 15:24
  • If you have reason to believe that the confounding variables effect on the dependent variable is different when treatment = 0 compared to treatment = 1, just divide your data into two subsets (one for treatment = 0, one for treatment = 1) and fit two separate models. Otherwise, the parameters for the confounding variables need to be shared somehow between treatment = 0 and treatment = 1; the formula you showed above is one possibility, there are others. Probably stats.stackexchange.com is a more suitable forum for this question. – Robert Dodier Jan 18 '18 at 20:58

0 Answers0