You weight the mean with the inverse squared of the errors. It may be a
root option right there. Else, you can take the mean of data/err/err (and
normalize to the sum of 1/err/err)
-- ________________________________________________________________________________ Tancredi Botto, phone: +1-617-253-9204 mobile: +1-978-490-4124 research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^On Sat, 7 Dec 2002, Peter Karpius wrote:
> Can anyone tell me how to exclude points based on an error bar limit from > a fit? Here is what I am doing: > > 1)Filling a 2D histogram of tdc top vs tdc bot > 2)Doing a FitSlicesY() on that histogram > 3)Fitting the means of the gaussians produced by FitSlicesY > > Now when I do FitSlicesY a few points have huge error bars. How can I > exclude them from the fit of the means? > > Thanks, > > Pete > ---------------------------------------------- > Pete Karpius > Graduate Research Assistant > Nuclear Physics Group > University of New Hampshire > phone: (603)862-1220 > FAX: (603)862-2998 > email: karpiusp@einstein.unh.edu > http://pubpages.unh.edu/~pkarpius/homepage.htm > ---------------------------------------------- >
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST