Re: Formula for Coefficient of Determination (Quadratic Regression)

From: Giobibo (giobibo_at_yahoo.com)
Date: 08/26/04 

Date: Thu, 26 Aug 2004 07:26:27 +0200

On Wed, 25 Aug 2004 15:35:18 -0400, Richard Ulrich
<Rich.Ulrich@comcast.net> wrote:

>Do you mean Y= a*X +b*X^2 + c ?
>
>Most statisticians consider that a linear regression because
>it is linear in the coefficients. Its R^2 is the CD.
>
>Many engineers (inter alia) don't call it linear, but it
>is still the simple case. Here is a post from last year,
>and my response -
>
>Is this a question about computing "R-squared", or
>about using the term "coefficient of determination"?

Dear Richard, thank you for your answer.

Yes, I mean Y= a*X +b*X^2 + c and want to compute R-squared.

II could not find any formula over the net.
This links helped me a little bit, but I have still a lot of dubts:
http://stat.tamu.edu/stat30x/notes/node47.html
http://www.medcalc.be/manual/mpage06-04c.php
http://www.cmh.edu/stats/ask/rsquared.asp

It should be R-squared = 1 - SSRes/SSTotal

where SSRes = Sum (Yesti - Ymean)^2
and SSTotal =Sum (Yi - Ymean)^2

Ymean is the mean of the samples, Yi is the i-sample, Yesti is the
predict i-sample.

Is this right?

Thank you for replying.
Gio

Relevant Pages

  • Re: advise on math course
    ... > Linear and Nonlinear Equations, Linear Least Squares, Eigenvalue ... mainly if you are going to work with engineers. ... > chain models, Markov decision processes, discrete-event simulation ... program to solve his optimization problem. ...
    (comp.theory)
  • Re: SW World opinions
    ... good luck in your mission ... I find engineers typically have real trouble with id people because they are ... quite uncomfortable in themselves about things outside the linear, ...
    (comp.cad.solidworks)