Indeterminacy in Small Region DSGE Model

Hello, I am trying to write a small region DSGE model as in Tamegawa (2013) “Constructing a Small Region DSGE Model” in Dynare. I have written the model in log-linear form, but when I run the mod file, the BK conditions are not satisfied as it says that there are 5 eigenvalues for 7 forward-looking variables. I understand the timing conventions are met (I am using the predetermined_variables command for specifying predet. variables and wrote the model equations without lagging such variables for one period). I have attached the mod file. Any help and suggestions with this issue are well appreciated. Thank you in advance.small_region.mod (2.5 KB)

Where does

  Y(+1)=Y+rn-pi(+1)+zd;

come from?

Hello Professor, thank you for your response. That equation represents the dynamic IS curve (Euler equation) rewritten in terms of the expectation in period t of Y in period t+1. Thus, Y and rn represent percentage deviations from steady state of national output and the monetary policy rate, respectively, pi is inflation, and zd is a demand shock.

Why did you rewrite it this way? And why is there no discounting in this equation?

Hello professor,
I did not have a specific reason to rewrite it that way. If I write in the standard form (in terms of expected output gap plus expected inflation) the results do not change. With regards to discounting, I’m not sure if you are referring specifically to the absence of the discount rate beta, which does appear in the NKPC or another parameter. That is one of three equations that describe the macroeconomy part (or the national economy) in the small region model, the other two are the NKPC and the monetary policy rule, as in the typical New Keynesian three equation model. This is the link to the paper, I hope it helps to clarify: https://www.hindawi.com/journals/isrn/2013/825862/