In an attempt to evaluate a potential U.S. National Sales tax rate, it was suggested on this forum before that I consider a difference-in-difference on state sales tax rates before and after they changed. How would I go about doing this? Would I do a difference-in-difference on each of the states sales taxes, and then aggregate the data? If so, how do I aggregate the data?

  • 2
    $\begingroup$ Could you add a reference to the place where this has been suggested to you? $\endgroup$ – user4239 May 19 '15 at 18:45
  • 1
    $\begingroup$ Well ... I know that they will lapidate me for saying this but I still believe that your regression approach is fine for your purpose. Really. $\endgroup$ – user4239 May 19 '15 at 20:41
  • 2
    $\begingroup$ @Howsikan nope, I will have to agree with FooBar. The relevant tool is DID. $\endgroup$ – VicAche May 19 '15 at 23:12
  • 2
    $\begingroup$ @Howsikan as FooBar tried to explain you, a regression quantifies a correlation, not causality. You cannot have predictive results with causality. $\endgroup$ – VicAche May 19 '15 at 23:17
  • 1
    $\begingroup$ try nber.org/WNE/lect_10_diffindiffs.pdf $\endgroup$ – VicAche May 19 '15 at 23:19

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.