Yes Johannes,you are right: we do not compute moments across different samples. I think we should if simul_replic>1. With Sebastien we decided not to do this just before the release of 4.3.0.
Best, Stéphane.
On 24/06/2012 22:40, Johannes Pfeifer wrote:
Dear Michel,
thanks for the answer. I am not really worrying about the binary-file. I rather have the impression, the replic option does not do what it is supposed to do. Am I right that regardless of simul_replic (or replic before 4.3) the simulated moments displayed by Dynare are actually always based on only one sample?
Best,
Johannes
--
Johannes Pfeifer
Haußerstr. 29
72076 Tübingen
Tel.: +49-(0)7071-6396184
Mobil.: +49-(0)170-6936820
Germany
Von: dev-bounces@dynare.org [mailto:dev-bounces@dynare.org] Im Auftrag von Michel Juillard Gesendet: Sonntag, 24. Juni 2012 21:53 An: List for Dynare developers Betreff: Re: [DynareDev] Help needed
Dear Johannes,
you are right. This code doesn't make much sense. It only creates several (replic) simulation samples in a binary file, but doesn't do anything with it. I had forgotten about the binary file. I doubt anybody ever used it.
Best
Michel
On 06/24/2012 07:08 PM, Johannes Pfeifer wrote:
Dear all,
can anyone tell me what is going in simult.m when replic>0?
The different replications are written into a file
fname = [M_.fname,'_simul'];
but this file has no valid file ending and is – as far as I can tell - never called or used later on.
Moreover, I cannot identify how the moments are computed across replications. After the last run
for i=1:replic
if ~isempty(M_.Sigma_e) oo_.exo_simul(:,i_exo_var) = randn(options_.periods,nxs)*chol_S; end y_ = simult_(y0,dr,oo_.exo_simul,order); % elimninating initial value y_ = y_(:,2:end); if replic > 1 fwrite(fh,y_,'float64'); end
end
the last simulation result y_ is saved into oo_.endo_simul which is then used by disp_moments.m to display the simulated moments. Hence, it looks to me as if only the last replication is used. What am I missing?
Best,
Johannes
Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev