Decimal precision

[Fab.valenti]

Dear all,

I have a, hopefully, pretty simple question. I have a parameter very close to zero (10^-6) and I'm trying to see how making it smaller and smaller affects the results of my model. However, changing its value to, for example, 10^-8, I obtain the same exact solution. This shouldn't be happening, so I suppose that this results from the way Dynare considers such values, i.e. that after "x" decimal digits my variable is approximated to zero. Is this the case? If yes, is there a way to go around it?

Thank you in advance.

[jpfeifer]

You must be more precise. Are you computing the steady state numerically and how to you judge that you get exactly the same solution?

[Fab.valenti]

I'm giving dynare the exact steady state values (that I computed separately). The parameter in question doesn't affect the steady state of the model, but influences the results that are under oo_.mean, or, better, changing this parameter affects the results in oo_.mean up to a certain point: when I give it values that are lower than 10^-6 (keeping everything else constant) the effect disappears. Am I missing something?

[jpfeifer]

In that case, I would need the mod-file.