Print

---

Now the blinding in the extraction of the ratio of BR's is in CVS.

The output tells you if it is on or off. Pay attention in switching it
off (or change the seed) in the MC fits.

Below you can find the log message I wrote in my commit as a description
of the blinding.

Cheers,

Daniele


===========================

Modified Files:
        VubFit.cc envfit.csh fitNtp.cc fitNtp.hh inputfiles.dat
        recoilAnalysis.cc scriptfit.csh settingfit.dat Log Message:

fitNtp:  implement the blinding

         - factor that multiplies the number of signal events (or eps_u as
	 you like)

         - the factor is extracted as a random number with uniform
         distribution


         Two parameters can be set by the user in settingfit.dat:

         1) seed (randomseed) = integer used as a seed in the randomic
                   extraction I chose the official random number (my
                   birthday)
                   -> ALREADY in settingfit.dat

         2) max interval of the uniform distribution (blindsize).
            I chose the official value (=0.4)


         Then at the moment the factor is a number in the interval
         (.6-1.4) with a uniform distribution


         I did some other changes.

         a) mx distributions for data are not shown with the blinding
            switched on (because of technical reasons) I will fix this
            in the future

         b) some new number are saved in the output file (results...dat) as
            the number of true events in vub, vcb and other categories for
            Mx<Mxcut. The corrected number of signal events (i.e. divided by
            the efficiencies is also saved). With the blinding swuitched
            on only this corrected number of signal events is shown

         c) the fit with the blinding on does not show the fit output
            (quite mode)

         d) all crucial mes fits are now saved in a ps file
            (...mes_misc..ps)