Page 1 of 1
Problem with the estimation of MSVAR examples
Posted: Fri Jan 13, 2012 5:17 am
by condor
When I try to run the MSVAR example files of Krolzig (1997) without changing anything, I receive the following error message:
"SX11. Identifier EMPTSM is Not Recognizable. Incorrect Option Field or Parameter Order?
>>>>VARMarginal(EMPTSM(<<<<
I am not sure if I am doing anything wrong...
Re: Problem with the estimation of MSVAR examples
Posted: Fri Jan 13, 2012 11:38 am
by TomDoan
That should now be MSVARPTSM rather than EMPTSM.
Re: Problem with the estimation of MSVAR examples
Posted: Tue Feb 28, 2012 3:48 am
by condor
Filardo (1994) example also yields error messages regarding EMSize and EMLagState:
## SX11. Identifier EMSIZE is Not Recognizable. Incorrect Option Field or Parameter Order?
>>>> do k=1,EMSize<<<<
## SX11. Identifier EMLAGSTATE is Not Recognizable. Incorrect Option Field or Parameter Order?
>>>>isEntry(EMLagState(<<<<
Re: Problem with the estimation of MSVAR examples
Posted: Wed Feb 29, 2012 10:33 am
by TomDoan
I just posted an updated version of the Filardo program.
Re: Problem with the estimation of MSVAR examples
Posted: Thu Mar 01, 2012 4:48 am
by condor
Hi Tom,
Thanks for the post, but it still gives the same error message. I think it has something to do with the following part:
Code: Select all
do time=start,end
compute %MSVARPMat(time)
compute thisEntry=%zeros(nstates,nstates)
do k=1,EMSize
compute thisEntry(EMLagState(k,1),EMLagState(k,2))+=EMPTSM(time)(k)
end do k
compute w11(time)=p(1,1)*(1-p(1,1))*(thisEntry(1,1)+thisEntry(2,1))
compute w22(time)=p(1,2)*(1-p(1,2))*(thisEntry(2,2)+thisEntry(1,2))
compute l11(time)=(thisEntry(1,1)- p(1,1)*(thisEntry(1,1)+thisEntry(2,1)))/w11(time)
compute l22(time)=(thisEntry(2,2)-(1-p(1,2))*(thisEntry(2,2)+thisEntry(1,2)))/w22(time)
end do time
I am not sure, but it feels like EMSize and EMLagState should be replaced with MSVARFilterSize and MSVARFilterLagState respectively...
Re: Problem with the estimation of MSVAR examples
Posted: Thu Mar 01, 2012 3:29 pm
by TomDoan
condor wrote:
I am not sure, but it feels like EMSize and EMLagState should be replaced with MSVARFilterSize and MSVARFilterLagState respectively...
Sorry. That's correct. I just posted a fully corrected version.
Re: Problem with the estimation of MSVAR examples
Posted: Tue Mar 06, 2012 6:00 am
by Aktar
it seems to be also the case in the Replication file for Krolzig's "International Business Cycles: Regime
Shifts in the Stochastic Process of Economic Growth.
Re: Problem with the estimation of MSVAR examples
Posted: Tue Mar 06, 2012 9:30 am
by condor
Aktar wrote:it seems to be also the case in the Replication file for Krolzig's "International Business Cycles: Regime Shifts in the Stochastic Process of Economic Growth.
Yes, but Tom has already proposed the remedy to this problem. Please see the second post under this thread:
http://www.estima.com/forum/viewtopic.p ... 1302#p4728
Re: Problem with the estimation of MSVAR examples
Posted: Wed Mar 07, 2012 1:03 pm
by TomDoan
Aktar wrote:it seems to be also the case in the Replication file for Krolzig's "International Business Cycles: Regime
Shifts in the Stochastic Process of Economic Growth.
It looks as if we never had those posted on the forum. Current versions are now at:
http://www.estima.com/forum/viewtopic.php?f=30&t=1362