in article [EMAIL PROTECTED], Matti Overmark
at [EMAIL PROTECTED] wrote on 6/18/01 7:10 AM:
> Hi group!
>
> I´m new to this group, so...just you know.
>
> I have fitted a 3 rd degree curve to a sample (least square method), and
> I want to compare this particular R2 with that of
> a (similarily) fitted 2 degree polynom.
>
Let SS3 = Sum of Squared Errors for 3rd degree curve fit
Let SS2 = Sum of Squared Errors for 2nd degree curve fit
Then the proportional reduction in error (PRE) for adding the cubic term
above and beyond the quadratic term is given by
PRE = (SS2 - SS3)/SS2
PRE
And then F = ---------
(1-PRE)/(n-4)
Which is distributed as F with 1 and n-4 df with the usual assumptions of
normal, identically-distributed, independent errors.
Gary McClelland
Univ of Colorado
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
