I am having trouble implementing the description in the Dynare manual for shock_decomposition. My shocks are named ea, eg etc. I thought I could use:
but I get the error message, which I really don’t understand.
“Starting Dynare (version 4.1.1).
Starting preprocessing of the model file …
ERROR: my_mod_file.mod:697.42-43: syntax error, unexpected NAME, expecting LABELS or PARAMETERS or SHOCKS”
Using the “labels” option works in principle, so I am really puzzled about how to map the manual syntax into code. I would appreciate any help - although in the worst case I would build my own plots on basis of oo_.shock_decomposition.
could you please post the mod-file so one can replicate the error.
Sure. Thanks for checking! I reproduced the same syntax problem in the baseline Smets-Wouters (2007, AER) model (see attached).
usmodel_dynare_forum.zip (22.2 KB)
Sorry, but the steady state file is missing.
unfortunately, the documentation of the syntax in the manual seems to be out of date. Try for example
to get a shock decomposition for the variable y.
Thanks for noticing the error in the reference manual, this will be fixed.
I’m trying to make sense of what the new manual says on this (on the options)
The new manual says the options as parameters=…, shocks=…, labels =…
How does one actually implement these options and what do they do? I tried to implement these options but I am getting syntax-type of error messages, so I cannot see the output of what these options do. I wish the manual would say a little bit more, or with some example.
Any help please? Thank you very much. I am new to this, and for clarity, please explain like you are explaining to a second grader, you will not insult my intelligence thank you in advance guys.
Actually there was a mistake in the reference manual: no option is currently implemented to the “shock_decomposition” command. You can only specify a list of endogenous for which to compute the decomposition.
New options should be implemented in Dynare next major release 4.2.
Ok Sebastien, thanks for clarifying. I spent a little time playing around with these options, I thought I was doing something wrong. Thanks for clarifying I was not in the wrong track.
For a version working in Dynare 4.2.5, see here: [Smets/Wouters (2007) in Dynare 4.2.5)