Page 1 of 1

no solution without a worthless variable

PostPosted: Sat Mar 12, 2016 6:56 pm
by shengan28
I created a deterministic model based on the Schorfheide fs2000a file. I removed his monetary injection variable and replaced it with a variable called ER for excess reserves in the bank's balance sheet in line 35. I also have a variable in the labor equilibrium equation I just call z in line 27. Without this variable, Dynare gives me the no solution or infinite solution warning. With the variable z, the program works, but seems to remove the impact on the labor market. My issue is that I am trying to make this file work without the worthless variable z. Please advise.

Re: no solution without a worthless variable

PostPosted: Tue Mar 15, 2016 8:18 am
by jpfeifer
You need to rethink your model. z is not a redundant variable. It only appears in this one equation. If you delete it, there are 14 equations but 13 endogenous variables, i.e. the model is overdetermined. This cannot be. Effectively, you have one equation to much. Only by adding a residual z to one equation can you make all equations hold at the same time.