I have recenlty run an old code on a new ROOT version. I do fitting with an “LL” option which is now, I think, replaced by a “WL” option. So… what “LL” option does now? Does it act like “L”? My fits results currenlty are very different and I am trying to find out what to do to reproduce results made with version 5.29
That’s an interesting issue.
The current ROOT “User’s Guide” (v5.26) mentions the “LL” option (but not the “WL” one) in the “Chapter 5. Fitting Histograms” in “The Fit Method” section:
http://root.cern.ch/download/doc/ROOTUsersGuideHTML/ch05.html#d5e4997
http://root.cern.ch/download/doc/5FittingHistograms.pdf
while “LL” is not present (but “WL” is there) in the current (ROOT 5.30 and newer):
http://root.cern.ch/root/html/TH1.html#TH1:Fit@1
though “LL” is still present (but not the “WL”) in ROOT 5.28:
http://root.cern.ch/root/html528/TH1.html#TH1:Fit
Just for the case something disappears completely one day: [quote]“LL” - An improved Log Likelihood fit in case of very low statistics and when bin contents are not integers. Do not use this option if bin contents are large (greater than 100).[/quote] [quote]“WL” - Use Loglikelihood method and bin contents are not integer, i.e. histogram is weighted (must have Sumw2() set)[/quote]
Some further explanations can be found in [url]Fitting histogram weighted with low stats “WL” option
Looking at my results I think it does the same what “WL” does, but I am not completely sure