Synth is a package for R that enables one to construct a synthetic control group.

It constructs a synthetic control group based both on predictors, and on the dependent variable.


synth constructs this synthetic control group by searching for a weighted combination of control units chosen to approximate the unit affected by the intervention in terms of characteristics that are predictive of the outcome.

However, in my case, I am trying to estimate a difference-in-differences model without any predictors used as controls.

Hence, I have only one variable (the total amount of accidents in a city, in my case). I need to find the combination of control-groups that would resemble the treatment group in term of this variable only.

Is it possible to implement it using the Synth package?

  • $\begingroup$ A possible solution seems to use the dependent variable in the role of a predictor as well. This way, that would be 100% relevant to explain itself, and the result would be equivalent to looking at the dependent variable only. Is that correct? $\endgroup$ – Sam Aug 31 '18 at 8:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.