The RATS Software Forum

The attached zip file has the replication files for Bernanke & Mihov(1998), "Measuring Monetary Policy", QJE, vol 113, no 3, 869-902 (monthly data calculations). This includes estimation of structural VAR's, Markov switching estimate of an SVAR and Monte Carlo integration of a just identified SVAR.

Thanks Tom. This may be in the file already, but I was wondering if there was any code for Figure III (p.899) in the paper which shows the Bernanke-Mihov overall stance of monetary policy indicator? (The bmmonte.prg file says Figure III in it, but produces IRFs not the composite indicator series.)

No. That's not in there. Since the Markov switching estimates from the paper seem not to be reproducible, I made no attempt to do anything derived from them.

Dear Tom:

When I run BMMARKOV.RPF. Markov Switching Model, pp 890-891
p1 is 0, Figure I has no lines.

Best regard
Hardmann

The Markov Switching model in that paper seemed to be a throwaway addition that was poorly documented (as is described in the comments). I don't know if we ever figured out a way to get that to work. As it is, ML ends up with basically just a single regime.