Page 1 of 1

Dynare vs. GAMS: can anyone explain?

PostPosted: Mon Sep 15, 2008 11:51 pm
by erezyerushalmi
Dear All,

I'm trying to understand how to solve the most simple, static model.

I can do it with GAMS, and I am trying to do the same with Dynare.

I think my problem lies with the fact that GAMS solves with Mixed Complementarity, while Dynare linearizes... ???

I must add that I don't (yet) know how to use Matlab.

However, this is basically have a system of 7 variables and 7 equations. Dynare finds a "spurious convergence", while I get a convergence in GAMS.
The initial values that I'm using are the GAMS convergence, which should be correct.

Here is the model.
1 agent with a Cobb-Douglas utility. Consumes X, Y
2 production sectors, using K, L in Cobb-Douglas. World is endowed with K=1, L=1.
Tax is my shock parameter, which is transfered to the Agent.

I'm attaching both Dynare and GAMS code, and a pdf file of the equations.

Hope someone there can give me direction.

Thanks, Erez