Indeterminacy Issue

[tung_1976]

Hi,
I have experienced an 'indeterminacy issue' from Dynare even I tried to replicate some papers . After reading some topics, i.e. Rank conditions, in this forum, it seems to me that to fix this problem we have to modify the model rather than randomly choose some parameter values giving us the unique solution. Is my understanding correct? But, I still wonder why I can't replicate some paper although I use the same parameter values. Thank you in advance for any suggestions.
Best,
Pisut

[jpfeifer]

Hi,
your understanding is not correct. Changing the model mostly relates to models written by yourself where everything else should be correct. If you are trying to replicate models from published papers, they almost surely run with the model specification and parametrization given in the paper (apart from typos). So if you experience indeterminacy issues with these models, there is something wrong with your implementation, most probably a conceptual mistake. The most common one would be taking the timing exactly as given in the papers. Note that Dynare dates predetermined states given in papers like K_t+1 in the time period they are determined, that is you have to put them into Dynare as K not K(+1) (See the user guide or the forum for more information). A wrong timing is the most common reason for indeterminacy issues when replicating models.
You may want to post the mod-file so that people can have a look at it.

[tung_1976]

Thank you!