Short and Long Run Identification in Bayesian VAR

[AhmedSahlool]

Dear Tom,

I hope this finds you well,

Would you kindly tell me if there is a code for estimating a Bayesian SVAR with short and long term restriction. I found this in a the following paper:

"Can External Shocks Explain the Asian Side of Global Imbalances? Lessons from a Structural VAR Model with Block Exogeneity"

Or if we could mix parts of different to estimate such model?

Thank you in advance.

Ahmed Sahloul

[TomDoan]

Dear Tom,

I hope this finds you well,

Would you kindly tell me if there is a code for estimating a Bayesian SVAR with short and long term restriction. I found this in a the following paper:

"Can External Shocks Explain the Asian Side of Global Imbalances? Lessons from a Structural VAR Model with Block Exogeneity"

Or if we could mix parts of different to estimate such model?

Thank you in advance.

Ahmed Sahloul

Have you contacted the authors? It sounds like they may have done some or all of it with RATS.

The short-and-long-run restriction of the type they are using doesn't add any complication---it's a just-identified model, so given the coefficients and covariance matrix, you can estimate them rather easily using the same ideas as in Bjornland and Leitemo. How they handle the "B" in the "BVAR" is a bit unclear. It certainly seems like they did a shrinkage prior since they're doing comparisons to OLS, but I didn't see where they reported the settings. The exogenous block, however, can be handled rather easily within a BVAR setup by doing a GENERAL prior with the other variables (nearly) zeroed out in the exogenous block's equations. See, for instance, the RUNTHEIL.RPF example.