Dynare forums

hi there,
I see some paper which build a stationary model directly( which means the system is built without trend and detrend by scaled ) normalize the steady state inflation to 1, is this normatization a must to ensure that the model is stationary or we can use the actual mean of inflation series?

There is no general answer to this. Normalizing steady state inflation to 1 is simply a way of avoiding many thorny issues. But depending on what you want to achieve/model, you might want to model steady state inflation explicitly. See the work of Guido Ascari for all the problems that occur when neglecting steady state inflation.

jpfeifer wrote:There is no general answer to this. Normalizing steady state inflation to 1 is simply a way of avoiding many thorny issues. But depending on what you want to achieve/model, you might want to model steady state inflation explicitly. See the work of Guido Ascari for all the problems that occur when neglecting steady state inflation.

thanks dear jpfeifer!