 |
 |
Hi all, > some problems with the links. I see only depleted plots. It' a matter of a D that needs to go in a H :) !!!! > > As far as the chi^2 problem I think you have too many bins (mes fits with > few events...). Reducing the number of bins and the range, putting all the > overflow and underflow in just one bin each could improve the situation. > a) in order to clarify this chisquare issue i have: --> Updated my page with chisquare plots (http://www.slac.stanford.edu/~asarti/recoil/enevelop) --> Put in that webpage usual comparison plots (in order to have a useful comparison with urs ones) b) linked in previous webpage the code I use to make chisquare calculations: this is exactly the same code Urs is using and I tought that the 'mes binning issue' was properly taken into account with that calculation. c) I'm going to try to eliminate this 'overflow' and underflow problem in order to improve the chisquare.... and find out a final decent comparison btw me and urs.... (.. actually most of the single comparison pots linked in urs webpage http://www.slac.stanford.edu/~ursl/talks/101502/ still have dummy links and a comparison is difficult..) ale > Another question. When you add the error from B-D systematics, are you > changing the mean value too? I see some chi^2's that are larger after > adding the syst error (see mm2 for B0). > > Daniele > > On Fri, 18 Oct 2002, Riccardo Faccini wrote: > > > I have run over the latest froduction of root files from Alessio and > > produced the best I could do for representation. > > It is all contained, including the explenation in > > http://www.slac.stanford.edu/~rfaccini/phys/vub/envelopes/envelopes.html > > > > The reference chi^2 do not lead often to the same conclusions as the plots > > posted the other day by Urs. I guess this is due to the binning, but it > > should be investigated because it lowers (if possible) my trust in the > > meaning of the chi^2. For istance the distribution of MM2 after lepton > > cuts in the depleted sample was 31 for 28 Dof in Urs's plots while it is > > 99 for 24 Dof in Alessio's ... > > > > This said that the plots are useful to see the impact of the systematics. > > In general the variation of the D decays BF are the contributions with > > the largest impact in all quantities. > > > > please let me know if there is anything I can do to improve the grafical > > representation or the explanation of the plots. > > ciao > > ric > > > > > >