In sci.stat.consult Gordon D. Pusch <[EMAIL PROTECTED]> wrote:
: Don't do it that way either --- it's notoriously ill-conditioned.
: It's better and more numerically stable to use the singular-value
: decomposition of 'A' to solve this problem.
It's NOT ill-conditioned unless the X'X matrix is ill conditioned,
Orthogonalization is certainly the method of choice, but with the
precision used in most programs, the Cholesky method will work just fine.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================