[R] Hessian from optim()

[Thomas Lumley]

On Tue, 21 Mar 2006, Ingmar Visser wrote:
> The optim help page says:
>
> hessian     Logical. Should a numerically differentiated Hessian matrix be
> returned?
>
> I interpret this as providing a finite differences approximation of the
> Hessian (possibly based on exact gradients?). Is that the case or is it a
> Hessian that results from the optimization process?
>

Yes, when it says "numerically differentiated" it means numerically 
differentiated.  It uses finite differences of the gradient (the analytic 
gradient if that was supplied, otherwise a finite difference 
approximation).

 	-thomas