Simulating OLG models with real data

This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location where you will have to reset your password.
Forum rules
This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location (https://forum.dynare.org) where you will have to reset your password.

Simulating OLG models with real data

Postby MK » Mon Sep 03, 2007 1:10 pm

Hello,

Would it be possible with Dynare to reproduce papers such as "The 1990s in Japan: A Lost Decade" by Hayashi and Prescott or "The Japanese Saving Rate Between 1960 and 2000: Productivity, Policy Changes, and Demographics" by Chen et al.?

Thanks in advance.
MK
 
Posts: 2
Joined: Mon Sep 03, 2007 12:58 pm

Postby reubenpjacob » Mon Sep 03, 2007 2:19 pm

yes it is possible. any DSGE model should work.
reubenpjacob
 
Posts: 133
Joined: Fri Oct 06, 2006 3:23 pm
Location: Reserve Bank of New Zealand

Postby MK » Mon Sep 03, 2007 5:44 pm

Okay, thank you very much. It is just, that I don't see how to confront the model with real data. Of course one could extend the program by hand-coding the relevant Matlab routines, but that is not what I am looking for. I admit I have not read all documentation, and generally should have researched the subject more, before I should ask this question.
I can paraphrase my question: is it possible, within the Dynare framework and without any further technicalities, to simulate models with real data, such as done in the said papers.

Again, thank you for your answer.
MK
 
Posts: 2
Joined: Mon Sep 03, 2007 12:58 pm

Postby reubenpjacob » Wed Sep 05, 2007 6:34 am

it is possible to estimate any DSGE model ( OLG or IFRA) models with this toolbox with real world data. however the fit of many of these models is not upto scratch since DSGE models, are essentially economic and not statistical models. we can estimate them as they have a statistical state space representation , and the likelihood can be evaluated and maximised as in a regression model. one needs to adjust the model, to fit the persistence of the data, so that the estimates makes some sense. one measure of the fit of the model would be the size of the std errs of the shock. the thing to be careful is that the model variables should correspond one-to-one with the data transformation for the observables.

cheers
reuben
reubenpjacob
 
Posts: 133
Joined: Fri Oct 06, 2006 3:23 pm
Location: Reserve Bank of New Zealand


Return to Dynare help

Who is online

Users browsing this forum: Google [Bot] and 8 guests