[Rd] constrained optimiser doesn't obey constraints (PR#411)

Thomas Lumley thomas@biostat.washington.edu
Fri, 4 Feb 2000 18:35:18 +0000 (GMT)

On Fri, 4 Feb 2000 thomas@biostat.washington.edu wrote:

> ---------- Forwarded message ----------
> Date: Fri, 4 Feb 2000 01:58:44 +0000 (GMT)
> From: Thomas Lumley <thomas@biostat.washington.edu>
> To: r-bugs@biostat.ku.dk
> Subject: constrained optimiser doesn't obey constraints
>  The constrained optimiser is calling the function with parameter values
> outside the constrained range

It seems to be happening in the numerical gradient computation. If the
current parameter value is on the boundary then one or other of the points
used for the finite difference will be over the boundary.

This explains why the values are over the boundary by a consistent small


r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch