Interpretation of IRF


I’m rather new at Dynare, and have looked through all the documentation I can find online, including Barillas et al.'s manual, Kamenik, and the standard manual itself (this is just to indicate that I am not asking without having first looked for the answer, given that this is such a simple question).

How does one interpret the IRF units? It’s unclear if the output is in standard deviations of the original variable, or in units of the original variable.

Any insight that people could give would be appreciated. I want to apologize again for asking such a simple question, I just cannot discern it from the code or any other searches.



The question has been answered. The existence of the “relative_irf” option indicates that the default IRF is in original units.

IRFs are in absolute deviations from steady state. See also [Dynare relative irf) and [Shock equations for model written in their non-linear form) and [A question about the IRF)