Help with MLE/Bayesian estimation DYNARE

[binhpham]

Dear senior dynarers,

I have a problem with my model + data. I propose the model as in mod file (+pdf description). I use quaterly gdp data quarterly in term of growth rate (% quarterly) so that I demeaned GDP growth (quarterly) from the series to get stationary data. To be more specific, I describe data construction:

1) Remove seasonal effect from quarterly GDP series using X12 in Gretl;
2) Compute ln_Y_obs = ln_Y - ln_Y(-1) to get growth rate;
3) Demean the series from step 2 to get final ln_Y_obs.

I then estimate the model but I always get beta = 1, and MLE/Bay indicates the model has no steady-state values.

I confuse with data used in estimation. Suppose I have only GDP growth rate (quarterly, y-o-y basis) I know I have divide it by 400 but how do I code in Dynare to use growth rate series correcly?

Please check my codes and guide me something to improve. I appreciate all your comments and time.

Thank you very much,

[jpfeifer]

Your are not handling parameter dependence correctly. See Pfeifer(2013): "A Guide to Specifying Observation Equations for the Estimation of DSGE Models" https://sites.google.com/site/pfeiferecon/Pfeifer_2013_Observation_Equations.pdf

[binhpham]

Thank you Professor Johannes. I made it works now.