Do we want to change saved structures? Can they be made more useful?

ZoltanRacz / MomentMatching.jl

This is a package to estimate parameters of macroeconomic models by moment matching methods, such as GMM or SMM.

MIT License

1 stars 0 forks source link

Do we want to change saved structures? Can they be made more useful? #51

Closed ZoltanRacz closed 9 months ago

ZoltanRacz commented 9 months ago

Some ideas:

a while ago we deleted number of iterations and another measure of convergence. should we bring back one of them?
should we put setup inside the saved results?

ZoltanRacz commented 9 months ago

eventually would be good to make it possible to pass any inputs for local optimization, in a way that doesn't make code messy. eg.abs_tol, rel_tol from Optim or the maximum time thing from NLopt

ZoltanRacz commented 9 months ago

"a while ago we deleted number of iterations and another measure of convergence. should we bring back one of them?" => no, NLopt doesn't support those things
"should we put setup inside the saved results?" => no, unnecessary extra work
"make it possible to pass any inputs for local optimization" => taking care of it