[EMAIL PROTECTED] (akhan) wrote:
"Could you give me some suggestions on how to test whether a model
fitted the dataset significantly?

Even if model A fits the data significantly better than model B, we
still don't know at what extent model A fits the data. Maybe both A
and B are far away from the truth."

This requires that "far away" be defined.  Assuming that one can
specify a tolerance (such as such-and-such mean squared error), and
one has a representative set of observations, then error may be
estimated via resampling and a decision reached by simple comparison.

-Will Dwinnell
http://will.dwinnell.com
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to