The RATS Software Forum

I have a question regarding the implementation of the Rigobon's VAR model for measuring the reaction of the monetary policy to asset prices based on the "Identification through heteroskedasticity" approach (Using the volatility regimes in order to achieve identification of the VAR). I have read anything I have found about VARs in RATS and still haven't found anything near Rigobon's modification.

I have the Gauss programs for the model, but as i have never worked with it, this does not help me a lot.

Any hint would be useful!

This is a very crude replication of Roberto Rigobon & Brian Sack, 2003. "Spillovers Across U.S. Financial Markets," NBER Working Papers 9640, which identifies the structural shocks in a VAR via a GARCH model.

Program file: Data file:

The original Rigobon and Sack paper on identification via multiple regimes used proprietary data. Lanne-Lutkepohl JMCB 2008 uses the same technique applied to public data.

dear
how to get the Impulse Responses mentioned in the paper?
thank you very very much

If you're talking about the "spillovers" paper (Simultaneous equations GARCH) that wouldn't be very difficult. However, this is a 2003 working paper that has never been published. I wrote the example four years ago, and just posted it here since we had a recent question. When the authors of a paper don't seem to think a paper is worth much, then I don't either.

I know that this is not a coding or RATS question, but I would appreciate any insight on this topic. How different are the methods used in Roberto Rigobon & Brian Sack, 2003. "Spillovers Across U.S. Financial Markets" from Rigobon and Sack (2004) "The impact of monetary policy on asset prices", JME, 51 or Lanne and Lutkepohl (2008) "Identifying monetary policy shocks via changes in volatility", JMCB, 40(6) ?

IRJ wrote:I know that this is not a coding or RATS question, but I would appreciate any insight on this topic. How different are the methods used in Roberto Rigobon & Brian Sack, 2003. "Spillovers Across U.S. Financial Markets" from Rigobon and Sack (2004) "The impact of monetary policy on asset prices", JME, 51 or Lanne and Lutkepohl (2008) "Identifying monetary policy shocks via changes in volatility", JMCB, 40(6) ?

Very different. The "spillover" paper is really a multivariate GARCH model with some (basically unstructured) VAR interpretation applied at the end. The identification by heteroscedasticity regimes is very much a structural VAR technique---the whole point is to estimate an SVAR which is not otherwise identified.

I am a newbie at the forum and econometrics as well. I am reading Rigobon (2004) paper but I do not know how to have the regression data? How did Rigobon run the model and what data is necessary to run the model? Do we need information on the common variables to both LHS variables? Could you explain that for me? Many thanks!

mntran wrote:I am a newbie at the forum and econometrics as well. I am reading Rigobon (2004) paper but I do not know how to have the regression data? How did Rigobon run the model and what data is necessary to run the model? Do we need information on the common variables to both LHS variables? Could you explain that for me? Many thanks!

As mentioned above, Rigobon and Sack(2004) used proprietary data. You would have to check with them about the source.

The Lanne-Lutkepohl JMCB 2008 paper is a similar idea and the authors provided the data set so I recommend you look at that instead.