Missing observations in the data?

Estimation
1

Hi all,
I am estimating the SW model with no missing observations using the mod and data mat files provided in Johannes’ GitHub.
This is what is generated:
dynare_estimation_init: There are missing observations in the data.
dynare_estimation_init: I am computing the moments for the endogenous prior only
dynare_estimation_init: on the observations after the last missing one, i.e. 1.

Could you help to confirm if that is a bug?\Best,
Jonathan

2

Dear Jonathan,
I cannot reproduce that message. How exactly did you obtain it?
Best,
Johannes

3

Dear Johannes,
thank you for your reactivity. I thought it was a known issue that is why I did not provided information.
I am using the public replication files of my paper “Central Bank Losses and Monetary Policy Rules: A DSGE Investigation” (International Review of Economics and Finance, Elsevier, Volume 61, Issue 1, Pages 289-303, May 2019) available here: http://www.jonathanbenchimol.com/data/research/codes/IREF2019.zip

When I wrote the paper and used Dynare 4.4.3, here is what I obtained from the very beginning:

Starting Dynare (version 4.4.3).
Starting preprocessing of the model file …
Found 48 equation(s).
Evaluating expressions…done
Computing static model derivatives:

  • order 1
    Computing dynamic model derivatives:
  • order 1
    Processing outputs …done
    Preprocessing completed.
    Starting MATLAB/Octave computing.
    Loading 248 observations from usdata0.xlsx

However, when I use the same mod and data files with Dynare 5.4, here is what I obtain:

Starting Dynare (version 5.4).
Calling Dynare with arguments: console
Starting preprocessing of the model file …
Found 48 equation(s).
Evaluating expressions…done
Computing static model derivatives (order 1).
Computing dynamic model derivatives (order 1).
Processing outputs …
done
Preprocessing completed.
dynare_estimation_init: There are missing observations in the data.
dynare_estimation_init: I am computing the moments for the endogenous prior only
dynare_estimation_init: on the observations after the last missing one, i.e. 1.

Thanks a lot for your help.
Best,
Jonathan

4

Is there a particular file I need to execute in the package?

1 Like
5

Any mod file with any data file (the mod file calls usdata0.xlsx but you can change, this does not change this phenomenon).
I also tried to convert these xls datafiles into mat files, same result.

6

Is this a Dynare bug or mode file typo?

7

Sorry for the delay. I can confirm this is a bug: endogenous prior: fix logic of missing observations check (!2153) · Merge requests · Dynare / dynare · GitLab
Note that it only affects the display of a warning message. Everything is computed as before.

1 Like
8

Thank you very much Johannes