MJDGGES and Steady State File

Dear all,

I am quite new to Dynare, using the 4.2 version, and I want to estimate a big model that is written in log-deviations. Before working on the big model, I wanted to estimate a small one, with only one agent and capital, no labor, and using a _steadystate.m file. As I only try to check whether the model - with a standard calibration - satisfies the Blanchard-Kahn conditions, I run onto the following error message:

??? Errors using ==> print_indo
MJDGGES returns the following error code: 3

Error in ==> check at 51
pring_info(info, options_.noprint);

Error in ==> simple at 90
check;

Error in ==> dynare at 132
evalin(‘base’, fname);

I can not figure if this error comes from 1) my steadystate file that does not send back the steady state values properly, 2) There is an issue with my model, which is unlikely since the non-linear version works with dynare or 3) I made a calculus mistake in my log-deviation, which would be shameful.

I join to this post my .mod file and my steady state file, that I took on this forum.

Thank you very much for your time,
simple_steadystate.m (430 Bytes)
simple.mod (271 Bytes)

Always put steady before check. The problem is that you did not program a correct steady state file. See [Steady state file)

Dear Sir,

Thank you very much for your answer. I used the steady state file you mentioned, and it works! I learnt a few things here, so thank you very much. I attach these files to this message, so other can use it, but there are a few minor issues:

It works perfectly fine but this little warning message that I don’t get either when executing the steady request:

[quote]Warning: Some of the parameters have no value (alpha, beta, sigma, gamma, rho, cs, ks, rs, ls, as, ws) when using steady. If these parameters are not initialized in a steadystate file, Dynare may not be able to solve the model…

In test_for_deep_parameters_calibration at 46
In steady at 33
In simple at 94
In dynare at 132[/quote]

Nonetheless, after executing the .mod file with dynare, I get all the parameters’ values in the M_.param vector. Isn’t that puzzling?

Once again, thank you very much for your time.

PS : what use to be my first problem:

EDIT : NOW SOLVED : I needed to put the values of my calibrated parameters into the steady state file. The first block of your program does not seem to work when the function is used by the .mod file, but works perfectly fine when I write it in the terminal. So I get the following error code:

[quote]??? Error using ==> beta
Not enough input arguments

Error in ==> simple_steadystate at 14
rstar = (1-beta)/beta;[/quote]

This can easily be overcome by putting the values of alpha, beta etc. directly in the steady state file, but it not supposed to be needed I guess, regarding with the first block of your code. I suppose that this could be a source of problems if I want to estimate the model…

RESOLUTION : This was a beginner mistake I guess, since I just learnt that alpha, beta etc. are MatLab built-in functions, so I need to use other parameters name. This remark may be useful for other people investing the same issue on this forum, so I do not delete it from this post.
simple_steadystate.m (1.25 KB)
simple.mod (421 Bytes)

You must initialize your parameters in the mod-file before the model-block. See e.g. the example1.mod in the Dynare example folder. Do not put the general parameter initialization in the steady state file.

Dear jpfeifer,

Sorry for my late answer, I had been very very busy, but your post was very informative. Thank you again!