Error in optim with method L-BFGS-B when hessian is TRUE

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Error in optim with method L-BFGS-B when hessian is TRUE

R help mailing list-2
Dear members,

I fought for several days against an error I was getting with optim in
L-BFGS-B. The error was produced because some parameters were outside
the limits defined by upper and lower.
After investigation, the error is not produced during the optimization
itself but during the calculation of the Hessian matrix which does not
take into account the upper and lower bounds.
It would be nice if this was stated in the optim help within the
L-BFGS-B method section.

Marc Girondot

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Error in optim with method L-BFGS-B when hessian is TRUE

R help mailing list-2
Sorry... it is already stated in the help, at the hessian section:

hessian
Only if argument hessian is true. A symmetric matrix giving an estimate
of the Hessian at the solution found. Note that this is the Hessian of
the unconstrained problem even if the box constraints are active.

So no problem...

Marc

Le 23/08/2020 ?? 11:19, Marc Girondot via R-help a ??crit??:

> Dear members,
>
> I fought for several days against an error I was getting with optim in
> L-BFGS-B. The error was produced because some parameters were outside
> the limits defined by upper and lower.
> After investigation, the error is not produced during the optimization
> itself but during the calculation of the Hessian matrix which does not
> take into account the upper and lower bounds.
> It would be nice if this was stated in the optim help within the
> L-BFGS-B method section.
>
> Marc Girondot
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.