Blanchard Kahn conditions, Rabanal, Rubio-Ramirez (2005)

Hi everyone,

I am new to dynare so there are probably several interrelated sources of confusion here.

I am trying to replicate Rabanal and Rubio-Ramirez, “Comparing New Keynesian models of the business cycle: A Bayesian approach”, JME (2005).

They have their model already linearized around the steady state, so this is what I used for the model equations.

I began by simulating the model using the means of their estimated posterior distributions for parameter values. This goes through.

I then try to estimate their model using, as they do, data on output, prices, wages and interest rate. Here is where I am stuck.

For the eigenvalues I get

EIGENVALUES:
         Modulus             Real        Imaginary

          0.4391           0.4391                0
             0.8              0.8                0
            0.85             0.85                0
               1                1                0
           1.011            1.011                0
           1.047            1.047                0
             Inf              Inf                0
There are 3 eigenvalue(s) larger than 1 in modulus 
for 3 forward-looking variable(s)
 
The rank condition is verified.

So it seems I have a unit root. Might this be the reason for the following problem?

Error in computing likelihood for initial parameter values
Error using print_info (line 40)
Blanchard Kahn conditions are not satisfied: no stable equilibrium

Help appreciated,

Sebastian
rrr_data_1.xls (37 KB)
BSP_1.mod (1.68 KB)

Your model is not stationary, because the price level is not stationary. If you are supposed to use the price level as opposed to inflation, use the diffuse_filter-option