You need to discuss this with your adviser. I was opposed to replicating that paper from the beginning.
Perfect, I prefer Bayesian estimation over ML as well. The key question, however, is, can I run parameter stability tests in this case. I can’t find any “parameter stability” testing references on this forum.
I want to emphasise, that I want to simplify and shorten the stability testing as much as possible, as I’m entering PhD area here and I’m only writing master’s thesis.
Instead of "Estimate of Set of Stable parameters”, developed by Inoue and Rossi (2011), I want to employ just the Andrews (1993) QLR stability test. Alternatively, if you know even simpler test, I would be glad to hear it.
So, my question is, could you suggest a stability test that uses Bayesian estimation.
Do you have a known break date?
No, I don’t, I want to test for the presence of breaks.
I haven’t studies that subject before. You should talk to your advisor.
Dear professor,
after one year of trying to run parameter stability test that you suggested me to abandon, I finally agreed with you, because the testing procedure is simply too complicated.
Now, I want to just run the estimation and then do the simplest stability test possible in Dynare, i.e. Chow test.
But running estimation files, that worked one year ago, now don’t work. I am attaching all three files. The error message I obtain is:
WARNING: in the ‘steady_state_model’ block, variable ‘Dataitmobs’ is not assigned a value
WARNING: in the ‘steady_state_model’ block, variable ‘Dataitinobs’ is not assigned a value
WARNING: in the ‘steady_state_model’ block, variable ‘Dataitcobs’ is not assigned a value
warning: Some of the parameters have no value (lne, lnz) when using steady. If these parameters are n
ot initialized in a steadystate file or a steady_state_model-block, Dynare may not be able to solve t
he model. Note that simul, perfect_foresight_setup, and perfect_foresight_solver do not automatically
call the steady state file.
You did not declare endogenous variables after the estimation/calib_smoother command.
error: Octave cannot create fig files!
error: called from
dyn_saveas at line 56 column 9
plot_priors at line 76 column 5
dynare_estimation_init at line 338 column 13
dynare_estimation_1 at line 109 column 93
dynare_estimation at line 118 column 5
driver at line 498 column 14
dynare at line 281 column 5
reestimate model.m (371 Bytes)
data1.mat (4.9 KB)
Model6.mod (4.0 KB)
In Octave, you cannot have graph_format=fig.
I deleted this option, but now Octave freezes if I run this commands:
addpath C:\dynare\5.5\matlab
dynare Model6
I waited for long time and tried running it with fresh and cold computer in the morning, but nothing seems to help. Can you run the file successfully on your machine?
Please try upgrading Dynare and Octave to see whether that fixes the issue.
I have a problem. I unistalled Octave 8.4 and installed Octave 10.1, however command
addpath C:\dynare\10.1\matlab
doesn’t work as I don’t have 10.1 folder. But I still have old folders with Octave versions 5, 6 and 7, printscreen:
On the other hand, using old two commands:
addpath C:\dynare\5.5\matlab
dynare Model6
returns this output:
This version of Octave is not supported. Consider installing version 8.3.0 of Octave
from www.octave.org, otherwise m files will be used instead of precompiled mex files and some
features, like solution of models approximated at third order, will not be available.
Error: opening the library ‘C:\dynare\5.5\mex\octave\win64\num_procs.mex’ failed (error 126): Th
e specified module could not be found.
The most recent Dynare version is 6.3. It is compatible with Octave 9.4, see
I installed both Dynare 6.3 and Octave 9.4 but nothing changes. After commands:
addpath C:\dynare\6.3\matlab
dynare Model6
Octave freezes.
Did you follow
Thanks. The line:
% stoch_simul(graph_format = fig);
is present in this file, in the attachment:
Model6.mod (4.0 KB)
I downloaded unstable version of Dynare and used the usual two commands:
addpath C:\dynare\7-unstable-2025-05-21-1913-42bf79cf\matlab
dynare Model6
but Octave again freezes…
Having that line commented out means it is not present. You never set the figure type to fig.
Perfect, I deleted that percentage sign. However, now Octave crashes (closes itself) automatically few seconds after I run these commands:
addpath C:\dynare\7-unstable-2025-05-21-1913-42bf79cf\matlab
dynare Model6
Model6.mod (4.0 KB)