model "works" in LEVELS, but steady state not found for LOGS

[Dynared]

Hello everyone,

I am trying to get started in Dynare. The problem I have been facing is that my model seems to be sensitive to whether I specify the variables in logs (exp^X) or in levels (X).

First, sometimes, certain parameterizations only work in logs, but not in levels.

Second, in other cases, the "levels" model finds a steady state in Dynare, yet when I change even 1 variable from levels to logs, Dynare no longer finds a steady state.

Are there one or two typical reasons for this?

Thanks a lot!

[jpfeifer]

The typical reason is that you did a mistake somewhere so that the two are not equivalent. Parameterization cannot matter (unless one parametrization results in negative steady states and you take the log of them).

The second point suggests that when you switch to logs, you did not adjust the initial values accordingly. See Section "4.4 Linearization vs. Log-linearization" in 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 for details.