Minimize function won't show me the hessian

Question

I minimized a function and need it's (inverse) Hessian for the standard errors.

the function gives me this for the (inverse) Hessian:

hess_inv: <5x5 LbfgsInvHessProduct with dtype=float64>

I expected a matrix here? I tried list(hess_inv) but that does not work.

P.S. Am I correct that the square root of the diagonalvalues equals the standard errors?

Possible duplicate of [Estimate of Inverse Hessian Using SciPy Minimization](https://stackoverflow.com/questions/42288740/estimate-of-inverse-hessian-using-scipy-minimization) — paisanco, Aug 26 '18 at 21:58
I see now that paisanco is right and this is a duplicate of another topic. However, I don't understand the solution there: I do not understand how to use the todense() function correctly. I tried to use toarray() instead but also got an error. I cannot comment on the other topic yet because I am a new user. If someone here can explain it, that will be appreciated very much — Rens, Aug 26 '18 at 22:21

score 0 · Answer 1 · edited Dec 28 '20 at 10:39

0

parameters.hess_inv will give you the inverse hessian matrix

edited Dec 28 '20 at 10:39

Tomer Shetah

answered Dec 28 '20 at 09:53

Xueli Zhang

1 Answers1