SBVAR estimation

[joff]

Probably this is a trivial question but is it possible to estimate "normal" SBVAR within MS-SBVAR procedure? There is a sbvar command but I am not really sure how to use it. For example, how should example code on wiki be modified to estimate SBVAR instead of MS-SBVAR? Thanks a lot for any clarification.

[jpfeifer]

You can find some examples here: https://github.com/DynareTeam/dynare/tree/adbecb6fc9ed26bfdc968c36595aeb78d2a878ba/tests/ms-sbvar. There are files without the Markov-Switching (ms_) statements.

[joff]

Thanks for a quick reply! Can you also please tell me is there a way to obtain error bands for the IRFs like in the Sims-Bvar setup (at least for the just identified model)?

[StephaneLhuissier]

Unfortunately, error bands are not available with this procedure.

[joff]

This is actually a follow-up question on the above. Are there any plans to improve the SBVAR interface in the near future? For example to introduce error bands and to allow combining Sims-Zha priors with exclusion restricitions? I understand that there are a lot of thing currently going on, but I would imagine these ones would also be hinghly appreciated by the Dynare user (well at least I would:).

[HoutanBastani]

There are currently no such plans.

If you'd like these changes to be made, you can contribute the code yourself through https://github.com/DynareTeam/dynare or ask your institution to contract the changes with Dynare.