Re: [Gretl-users] MLE Failure in Local Whittle Estimation
On Mon, 15 Mar 2010, savos schmagges wrote:
i have tried to re-consturct gretl's calculation code for the
local whittle estimation via the mle-command and i am getting
very often the failure:
---
Using numerical derivatives
Tolerance = 1.81899e-012
failed to invert OPG matrix GG'
---
Has anyone an idea how i can solve the problem or avoid it?
Just a quick suggestion for now: trying giving the --hessian
option to mle:
end mle --hessian
It may be that computing the numerical hessian is more stable
than computing the variance via the OPG.
Allin Cottrell