See below!
On Fri, 5 Jul 2002, Daniele del Re wrote:
>
> Hi all,
>
> next check... I removed the kaon veto at all. The results are in the same
> direction... almost
>
> Q=0
> ---
>
> All BRBR = 0.020486 +- 0.0048398(stat) +- 0.00141174(MC stat)
> B0 BRBR = 0.00843454 +- 0.00786974(stat) +- 0.00232202(MC stat)
> Bch BRBR = 0.0284635 +- 0.00617635(stat) +- 0.00195309(MC stat)
>
> incredibly compatible results. Gap B0 Bch unchanged!!!
>
>
> Q=1
> ---
>
> here something happens
>
> All BRBR = 0.0285778 +- 0.00541327(stat) +- 0.00159545(MC stat)
> B0 BRBR = 0.0221152 +- 0.00878295(stat) +- 0.00263773(MC stat)
> Bch BRBR = 0.0330851 +- 0.00687407(stat) +- 0.00206717(MC stat)
>
> B0 shifts up. All too, as a consequence.
>
>
>
> HERE the chi square
>
> Q=0
> ---
>
> All chisq = 1.26608
> B0 chisq = 1.55029
> Bch chisq = 2.47945
>
>
> Q=1
> ---
>
> All chisq = 2.30101
> B0 chisq = 0.588285
> Bch chisq = 3.1975
>
>
> And the fits:
>
> http://www.slac.stanford.edu/~daniele/vub/
>
> Q0
>
> dataB0FINALnokselefitresults_nocat.eps B0
> dataBchFINALnokselefitresults_nocat.eps Bch
> dataFINALnokselefitresults_nocat.eps All
>
> Q1
>
> datach101B0FINALnokselefitresults_nocat.eps B0
> datach101BchFINALnokselefitresults_nocat.eps Bch
> datach101FINALnokselefitresults_nocat.eps All
>
>
> Bch for Q1 seems to be pretty bad. Looking at the chisquare and at the
> plots Q0 seems to be better.
Please stop making this biased and baseless statements. Both (Q0 and
have chi**2/ndf>2.0 (correct?!) and hence both yielding Pchi**2=0 to very
good approximation.
The word "better" makes no sense in this context ... both do not fulfill
the chi**2 fit hypotheses and apparently by looking at the plots both
suffer from the same problem! Therefore, if your errors are correct both
(Q0 and Q1) indicate that there might be also a problem in BCH
... another one .. just to much for my tast!
>The point is the usual one: which is the
> sistematics that takes into account this effect?
>
> - tracking
>
> - actually, even if we have no kaon veto in, we could have still problems
> of kaon ID since we are using it in reconstructing Mx.
>
> What about removing at all kaon ID (always pion mass) and see what
> happens?
>
>
> Daniele
>
>
>
|