Identification Error

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:

==== 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)
dataset_info, prior_exist, name_tex,1,parameters,bounds);

Error in KH_Estim (line 528)

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


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.

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??

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.

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.

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??

Hi Prof. Pfeifer,
Any recommendation, pls.

  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.

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