Stability Mapping (GSA)

[rickardo]

Hi all,

I am attempting to conduct stability mapping of a non-estimated DSGE model and am having some difficulty with

"dynare_sensitivity" as the code. The dynare files say you can run it without estimation however upon entering the options for stability mapping I receive errors relating to estimation. I am hoping someone has some insight as to how to conduct this. I am just after some stability mapping of 1 or 2 key parameters.

Regards

[jpfeifer]

Without a mod-file it is hard to tell. The thing is that without an estimation command you still need to describe the estimated parameters and var_obs. Moreover, if you do not specify an estimation command you need to provide the stability mapping command direclty with the necessary options. That is, pass the required options you would pass to estimation directly to GSA.

[rickardo]

I have provided the code.

I am able to run the MC simulations and am being estimated a % for which values are stable / unstable

However seem to be unable to obtain any graphs such as those suggested in the Ratto paper - basically I just want to compare 2 different policy rules and the stable values of the parameter "phi_spr". I have made the distribution normal for simplicity as the model itself is not estimated.

Should I be using the values "istable" and "iindeterminate" in the .mat file produced?

I was wondering if anyone had any ideas?

Regards

[jpfeifer]

How about deleting

Code: Select all

options_.nograph=1;