Re: comparing slopes for significant difference

From: Ray Koopman (koopman_at_sfu.ca)
Date: 03/11/05 

Date: 11 Mar 2005 12:57:43 -0800

Ray Koopman wrote:
> [...]
>
> Try fitting five zero-intercept slope-only models: one for each of
the
> four people separately, and the fifth for all the data jointly. Then
>
> (RSS5 - (RSS1 + RSS2 + RSS3 + RSS4))/3
> --------------------------------------,
> RSS5/(N-4)
>
> where RSSi is the Residual Sum of Squares from the ith analysis and N
> is the total number of observations, should be distributed as
F(3,N-4)
> if all four people have the same true slope.

Correction: the denominator should be (RSS1 + RSS2 + RSS3 + RSS4)/(N-4).

Relevant Pages

  • Re: comparing slopes for significant difference
    ... Ray Koopman wrote: ... > where RSSi is the Residual Sum of Squares from the ith analysis and N ...
    (sci.stat.math)
  • Re: Nonlinear Least-Squares curve-fitting
    ... If you are finding approximately the same residual sum of squares for ... or what seem to be multiple minima... ... the same residual sum of squares... ... a large number of data points; therefore, if the least squares function ...
    (sci.stat.math)
  • Re: Preconditioning and weighted least squares
    ... Ray Koopman wrote: ... > squares, and the answer will depend on P. ... I've seen another popular type of preconditioner is to transform ... This one looks statistically OK to me (ie. we are not weighting the ...
    (sci.math.num-analysis)
  • Request Paper
    ... Manuel Melgosa, Rafael Huertas, and Roy S. Berns, "Performance of ... recent advanced color-difference formulas using the standardized ... residual sum of squares index", ...
    (sci.engr.color)