- My impression is that it should be
r_obs=log(r)*4;
as r
in the model is the quarterly gross interest rate but it is the annual net rate in the data.
- For inflation: monthly levels of what exactly were multiplied. Please describe exactly what you are doing.
Perfect, I will correct the interest rate equation.
Regarding inflation, the monthly data are HICP, from Eurostat:
Inflation Eurostat 1996-2022.xlsx (266.2 KB)
Up to first order, quarterly net inflation is the sum of monthly net inflation. Thus, you could use in the data (1.5 + 0.6 + 0.2)/100
as pi_obs
and then match
pi_obs=log(pi)
Done, however now I obtain error:
䏂ERROR: Model.mod: line 54, cols 1-6: Unknown symbol: pi_obs
ERROR: Model.mod: line 55, cols 1-5: Unknown symbol: r_obs
Model.mod (3.1 KB)
Data.xlsx (19.4 KB)
Try
Modelred3.mod (3.1 KB)
Data.xlsx (19.4 KB)
Many thanks, I obtain some steady-state results. However, everything freezes after the first window (Priors) pops up.
Is that again the Octave graphics bug?
Yes, it appears so. However, this time I’m using the unstable version of Dynare.
The Dynare version will not affect how Octave handles Intel Graphics drivers.
I see, so how would I solve the bug problem?
Most probably you won’t be able to do that unless you can get Intel to fix their driver.
But earlier (post 64/131), using your advice, I downloaded the latest unstable version of Dynare and Octave didn’t freeze, all 10 windows popped up normally. Perhaps changing version of Octave and Dynare would solve the problem again… (I think you recommended me Octave 8.1 at some point)?
One question. When I encountered the Octave not responding/freezing earlier, you advised me to add graph_format = pdf as an option to stoch_simul.
Now, encountering the same not responding problem after first graph pops up, can I add this option to estimation command? Doing so, prevents freezing, although an error message appears:
ESTIMATION_CHECKS: There was an error in computing the likelihood for initial param
eter values.
ESTIMATION_CHECKS: If this is not a problem with the setting of options (check the
error message below),
ESTIMATION_CHECKS: you should try using the calibrated version of the model as star
ting values. To do
ESTIMATION_CHECKS: this, add an empty estimated_params_init-block with use_calibrat
ion option immediately before the estimation
ESTIMATION_CHECKS: command (and after the estimated_params-block so that it does no
t get overwritten):
error: Fortran procedure terminated by call to XERBLA
I’m running Octave 8.3 with the latest unstable version of Dynare.
The current unstable version of Dynare only supports Octave 8.4:
Okay, using latest unstable version Dynare 7 and Octave 8.4 I get into a loop that produces infinite output. Can I add option graph_format = pdf or something else to estimation command to prevent freezing/not responding of Octave?
Model.mod (3.1 KB)
What type of infinite output?
Using option graph_format = pdf to estimation command, I get into a loop that returns output of lambda values, as per attachment.
That is not an infinite loop, but rather mode-finding, which can take some time.
But is using option “graph_format = fig” to estimation command correct?