Re: retain the fit and remove the curve



Another approach might be to use "Generate-M File" in the "File" menu. You
could then see how the fit is calculated and plotted. You would need to
modify the generated file to plot only the fit.

"Pekka Kumpulainen" <pekka.nospam.kumpulainen@xxxxxxxxxxxxx> wrote in
message news:gd7cja$f8v$1@xxxxxxxxxxxxxxxxxxxxx

"S N RAY" <suryanarayana_gs@xxxxxxxxx> wrote in message
<gd72b6$jur$1@xxxxxxxxxxxxxxxxxx>...

A plot is made in Matlab using plot() command of some data and then using
the Basic curve fitting GUI in the graph a curve of some order of
polynomial is fitted to the curve.Now I wish to remove the original curve
retaining only the fit. I tried 'imsubtract' but to no help. How can this
be done ?

h = plot(.....); this will give you the handle to the line of the data.
The logical approach would be to delete the original line after the fit:
delete(h)
But for some reason this will also remove the fit (R2008b) (I really hate
these GUIs and the way MATLAB is more and more automagically doing stupid
things)

Anyway, this will work as a quick workaround:
set(h,'Visible','off') % hide the data
legend off % remove the legend



.



Relevant Pages

  • nokey causing not plotting a function?
    ... I'm trying to plot two set of points from data files along with two fit curves for this data. ... Strangely, gnuplot misses one of the function, especially when "set nokey" is used. ...
    (comp.graphics.apps.gnuplot)
  • Re: Possible bug related with fit command and line style color
    ... do a fit and then try to plot a graph with customized line color, ... graph is shown. ... before the first fit command, ...
    (comp.graphics.apps.gnuplot)
  • Re: Least Square fitting
    ... I plot ... the data and fit a "rough" curve to it and found that the intercept of ...
    (sci.math)
  • Re: Limits to a fitting curve
    ... fit f'test1.DAT' using 1:2 via a,b,c,d ... I don´t need the curve to cover the whole 600-1000, only the area of the measured data should be covered. ... Just change your plot command to ... I don't know if it would make sense to allow a range specification for every function that is plotted and then take the range that includes all 'subranges' to plot the functions: ...
    (comp.graphics.apps.gnuplot)
  • Re: Interpreting Q-Q plot
    ... but instead of evaluating the goodness of fit ... line) for a GoF? ... You could try a PP or SP plot instead: see Michael The stabilized probability plot. ...
    (sci.stat.edu)