Page 1 of 1

Trivial question: Where are my results (octave)

PostPosted: Sun Mar 15, 2015 10:20 am
by ma'am
hi there,

I am new to Dynare and just started programming a DSGE model.

I am using Octave 3.8.1 with Dynare 4.4.3 on a Mac -- seems to work fine so far.
However, I did an estimation but I can´t find the results -- it gives me the IRFs as pop ups, but where is the data?

Is there anything like the oo_. files from Matlab?

Thanks!

Re: Trivial question: Where are my results (octave)

PostPosted: Sun Mar 15, 2015 11:26 am
by ptr
See response in the forum: viewtopic.php?f=1&t=4689&start=0

Re: Trivial question: Where are my results (octave)

PostPosted: Sun Mar 15, 2015 11:37 am
by ma'am
ptr wrote:See response in the forum: viewtopic.php?f=1&t=4689&start=0


Thanks for your reply, but I´m not sure if that is what I am looking for. It does give me the IRFs already.
Problem is that I am doing an estimation using the MH algorithm and that should get me values for my var, right?
But there are not any data files in my work directory.

I want to use the model to estimate the output gap.

Re: Trivial question: Where are my results (octave)

PostPosted: Sun Mar 15, 2015 5:39 pm
by jpfeifer
What exactly are you looking for. Dynare saves all results in the oo_ structure in the workspace. Which fields contain which results is documented in the manual. It seems you are looking for the smoother results.

The structures created by Dynare are subsequently stored in the _results.mat file on your harddisk. Note that saving by Dynare requires the mod-file to run without crashing due to an error.

Re: Trivial question: Where are my results (octave)

PostPosted: Sun Mar 15, 2015 10:06 pm
by ma'am
Thanks for the advice.
i´ve got it running now.

Unfortunately, the smoothed variables it gives me are not at all satisfying... output gap always negative around -0.2.

I attach my model and data file here.


i´m thankful for any advice.

Re: Trivial question: Where are my results (octave)

PostPosted: Mon Mar 16, 2015 10:19 am
by jpfeifer
You estimate sigma-parameters that are not in the model and your inflation rate has a strong seasonal pattern. Moreover, your observation equation seems to be wrong, perhaps by a factor of 100. See
Pfeifer(2013): "A Guide to Specifying Observation Equations for the Estimation of DSGE Models" at https://sites.google.com/site/pfeiferecon/Pfeifer_2013_Observation_Equations.pdf