tlumley at u.washington.edu
Thu Jul 10 17:37:04 CEST 2003
On Wed, 9 Jul 2003, Spencer Graves wrote:
> I haven't used optim enough to know, but simular functions ostensibly
> with box constraints would still test values outside the box and quite
> if it got an NA or possibly Inf. I've had good luck transforming the
> inputs to remove constraints. For example, if you need x > 0, then
optim() doesn't test values outside the box (even using finite difference
approximations for derivatives), but it does test values on the boundary.
Box constraints are for x>=0, which you can't get with transformations.
For x>0 transformations may well be indicated.
More information about the R-help