Alan-
Thanks for the help. I've done the math (I can do it manually for each
line if necessary), it's hacking the GX that's the problem because my
programming is zilch. One of these days I'll get around to learning some
C. Most of the tweaks I want are so small though, that it's more time
efficient to find a work around (like doing it manually).
-Jeff
At 02:28 PM 11/14/00 , you wrote:
>Jeff,
>
>There are (fairly) simple relationships to give you the best fitting
>slope, the best intercept and the variances and hence standard
>deviations of both of those, plus the cooefficient of correlation. You
>can even go and calculate chi-squared, which will tell you something
>about whether your fit explains everything or there are other factors.
>All this stuff is in standard texts on elementary experimental
>statistics, more or less at freshman physics level (or maybe sophomore
>if it's a backwater college).
>
>The simple statistics assume that there are uncertainties in y (the
>measured magnetic field, say) but not in x (the distance along profile,
>say). It gets messy if both measurements are uncertain, especially if
>the errors are correlated (as in Rubidium-Strontium isochrons)
>
>You need to go through the data set and accumulate sum of X, sum of
>X-squared, sum of Y, sum of Y-squared, sum of XY, and of course N. You
>need all of those to get the best slope and intercept, so they'll all be
>in the GX already. It's a relatively short step from there to the
>uncertainties in slope and intercept, and the coefficient of
>correlation, so hacking the GX shouldn't be hard, if you're into hacking
>GX's.
>
>The math doesn't e-mail well. I can fax it if you can't find it in the
>ORNL library, but it'll be there for sure. Look for REGRESSION and
>CORRELATION.
>
>Regards
>
>Alan Reid
>-----------------
>
>Jeff Gamey wrote:
> >
> > I can run linear regression to get the slope and intercept of my data, but
> > I'd like a measure of error of fit to that line. Does anyone have a way to
> > do that apart from (a) manually inputting the slope and intercept into a
> > math expression for error of fit, then (b) calculating the std dev of the
> > error?
> >
> > Having it included in the linear regression GX would be ideal. If someone
> > wants to talk me through it, I'll try to modify it myself.
> >
> > *******************
> > * Jeff Gamey (4tj)
> > * Environmental Sciences Division [EMAIL PROTECTED]
> > * Oak Ridge National Laboratory (865) 574-6316 tel
> > * POBox 2008, Bldg 1505, MS6038 (865) 576-8646 fax
> > * Bethel Valley Rd., Oak Ridge, TN (865) 599-0820 cell
> > * 37831-6038
> >
> > _______________________________________________________
> > More mailing list info http://www.geosoft.com/support/listserv/index.html
>
>--
>Alan B Reid PhD
>Reid Geophysics
>49 Carr Bridge Drive
>Leeds LS16 7LB
>UK
>
>Phone: +(44)-113-261-0252
>Fax: +(44)-113-293-0062
>e-mail: mailto:[EMAIL PROTECTED]
>website: http://www.reid-geophys.co.uk
>_______________________________________________________
>More mailing list info http://www.geosoft.com/support/listserv/index.html
*******************
* Jeff Gamey (4tj)
* Environmental Sciences Division [EMAIL PROTECTED]
* Oak Ridge National Laboratory (865) 574-6316 tel
* POBox 2008, Bldg 1505, MS6038 (865) 576-8646 fax
* Bethel Valley Rd., Oak Ridge, TN (865) 599-0820 cell
* 37831-6038
_______________________________________________________
More mailing list info http://www.geosoft.com/support/listserv/index.html
