Identification Error

Identification/Sensitivity
1

Dear All,

I’m new here and please apologize me if my English confusing you all since I have poor English.
I downloaded and run the code with his data which the link to download below (DSGE models for developing economies: an application to Morocco) and found many errors but I can’t solve it.
Please check and help me solve this problem.

This is the link: https://maitlahcen.github.io/research

Error
"
==== Identification analysis ====

Testing prior mean
Error using identification_analysis (line 94)
There are NaN’s in the theoretical moments: make sure that for non-stationary models
stationary transformations of non-stationary observables are used for checking
identification. [TIP: use first differences].

Error in dynare_identification (line 308)
identification_analysis(params,indx,indexo,options_ident,dataset_,
dataset_info, prior_exist, name_tex,1,parameters,bounds);

Error in KH_Estim (line 528)
dynare_identification(options_ident);

Error in dynare (line 223)
evalin(‘base’,fname) ;
"

Regards,

2

As the message says, you cannot run the identification-command on varobs that have a unit root. You would need to define growth rates of the variables and run it on them.

3

Hi Johannes,
Since i’m new to dynare, would you mind to show me how to do this thing or any link or book that I can refer to to understand more about this??
Thanks

4

The model in the paper is log-linearized around its steady state, and the data provided with the paper have a unit root. I think you should transform the data first.

5

Looking at the mod-file, it is puzzling that pretty much all variables have a unit root. That usually indicate a serious bug in the model. So I would not proceed with data transformation, but rather check the model.

6

Hello all,
How can this problem be solved??
Is there any link or documents that I can refer to, to improve my understanding of such problem??
Thanks

7

Hi Prof. Pfeifer,
Any recommendation, pls.
Thanks

8
  1. Please allow sufficient time for answers.
  2. There is some error in implemented model. It will be hard to find. You need to check all derivations and equations. That’s all I can recommend. It is complicated by the fact that it is not your own model.
9

Hello prof…
Thanks for your recommendation.
Have a good day