Page 1 of 1

Dynare Theoretical moments Nan

PostPosted: Tue May 22, 2012 9:27 am
by DaniilN
I tried to write a code of a model that I found in Sosunov,Zamulin 2007 "Monetary Policy in an Economy sick with Dutch Disease". Unfortunately, I got infinite moments of some variables( 5).
Please, look at my code and I will appreciate any suggestions

Thank you in advance

Re: Dynare Theoretical moments Nan

PostPosted: Wed May 23, 2012 9:18 am
by kyri82
your model has a unit root - see the eigenvalues. That can be either because either you did not linearise the model correctly, or you set some parametres to values that generate a unit root. After a very quick look, I saw that you have:
- r=1. are you sure this is fine?
- Your exogenous process is: phooil(+1)=pho*phooil+oilshock; That means that the oilshock is an expected shock. Is that what you want?

Other things: What type of utility function do you have? Do you have investment adjustment costs..? Is it an open/close economy model?

Re: Dynare Theoretical moments Nan

PostPosted: Wed May 23, 2012 1:59 pm
by DaniilN
Kyri82, thank you for your response.
You know, I tried to play with "r" this parameter but nothing changed. There is no explanation in the article how to define it. Firstly, I thought that "r" is a variable but there is no any equation and no superscript t with it. So, i decided to denote it as a parameter.

Talking about phooil(+) and etc. I have already changed it, nothing changes too in my model.
It is an open model. As I understood there is no any adjustment costs but probably I am not right. http://www.cefir.ru/papers/WP101.pdf - this is the article page 6 -description of the model and basic equations

Re: Dynare Theoretical moments Nan

PostPosted: Wed May 23, 2012 2:19 pm
by kyri82
It is a small open-economy model and r is the interest rate and is exogenous thus it is constant - equal to the world interest rate. There are several ways to closing these models - check "Closing small open-economy models" by Schmitt-Uribe and someone else - and one is to assume an endogenous disc. factor as the author does in this case. Anyway, I am not sure what the problem is but make sure you put the equations in correctly. The author gives you the log-linear model so it should be ok. But my guess it is a parametre. What other parametres are you not sure what value he gives in the original paper?