From: Ethan A M. <sf...@us...> - 2011-05-05 19:36:09
|
On Thursday, May 05, 2011 12:01:53 am Bastian Märkisch wrote: > > I may be missing something, but why not just report both > > absolute and relative errors? The user is free to pick whichever > > is relevant to the particular case at hand. No extra options needed. > > > > Ethan > > I think you just proved that the name may be misleading ;). This is > about the interpretation of data errors (input), and the resulting > scaling with FIT_STDFIT of reported variable errors (output). This is > not about reporting relative errors as in a_err/a. From Thomas Mattison's mail that you linked to http://thread.gmane.org/gmane.comp.graphics.gnuplot.devel/3737/focus=3740 Gnuplot already does the right calculation internally, it just doesn't report it. It reports only a non-standard, fluctuation-sensitive error (although I will grant that it is possible in most cases to recover the standard error from the fit log by some simple hand calculations). I took this to mean that the issue is ultimately one of reporting. Right now we report errors calculated under one set of assumption; he is asking that we also report errors calculated under a different set of assumption. My question is why not just report both? The current values become a set of "FIT_*_SCALED_*" output parameters and the new ones become a parallel set of "FIT_*_UNSCALED_*" parameters. But I admit to saying this without having thought deeply about the underlying issue. Just ignore me if I'm not making sense :-) > Not only is scaling the errors wrong in certain cases, but these error > values are also saved to variables for further processing, e.g. in > labels. Since, the user should be able to select. With `set fit abs` > gnuplot reports the same errors as e.g. CERN Minuit. > > Any other ideas about the names? weights|errors? scaling|noscaling? > weights|real? > > Bastian > |
From: Bastian M. <bma...@we...> - 2011-05-05 19:48:39
|
Am 05.05.2011 21:32, schrieb Ethan A Merritt: > On Thursday, May 05, 2011 12:01:53 am Bastian Märkisch wrote: >>> I may be missing something, but why not just report both >>> absolute and relative errors? The user is free to pick whichever >>> is relevant to the particular case at hand. No extra options needed. >>> >>> Ethan >> >> I think you just proved that the name may be misleading ;). This is >> about the interpretation of data errors (input), and the resulting >> scaling with FIT_STDFIT of reported variable errors (output). This is >> not about reporting relative errors as in a_err/a. > > From Thomas Mattison's mail that you linked to > http://thread.gmane.org/gmane.comp.graphics.gnuplot.devel/3737/focus=3740 > > Gnuplot already does the right calculation internally, it just doesn't > report it. It reports only a non-standard, fluctuation-sensitive error > (although I will grant that it is possible in most cases to recover the > standard error from the fit log by some simple hand calculations). > > I took this to mean that the issue is ultimately one of reporting. > Right now we report errors calculated under one set of assumption; > he is asking that we also report errors calculated under a different > set of assumption. My question is why not just report both? > The current values become a set of "FIT_*_SCALED_*" output parameters > and the new ones become a parallel set of "FIT_*_UNSCALED_*" parameters. > > But I admit to saying this without having thought deeply about the > underlying issue. Just ignore me if I'm not making sense :-) > To me, adding a user option still seems to be a good solution, since there's also the issue of error variables ('set fit errorvariables'). I am not sure gnuplot already had that feature back then. Imho, data errors are either one or the other. >> Not only is scaling the errors wrong in certain cases, but these error >> values are also saved to variables for further processing, e.g. in >> labels. Since, the user should be able to select. With `set fit abs` >> gnuplot reports the same errors as e.g. CERN Minuit. >> >> Any other ideas about the names? weights|errors? scaling|noscaling? >> weights|real? >> >> Bastian |