Suppose I have a VAR model that relates income, consumption and investment. And suppose I want the parameters to vary by the political party in power: A, B or C.

As far as I know, Markov Switching VAR or MS VAR treats regimes as unobservable and estimates which regime is in which of the time periods from the data. Is there any variant of MS VAR where I can pre-specify the regimes: that is, provide which regime is there in which time period. In my case, I want to treat the party in power as the regime.

Any concise resource (notes, slides etc.) to read about its theory and also how to implement it in R/ Python/ Stata? If MS VAR is not applicable here, then any other suitable tachnique?

  • $\begingroup$ it seems like you just want to split the series by the party in power and run three different VARs. $\endgroup$
    – mark leeds
    Aug 31, 2023 at 0:27


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.