For the point discussed below In short the fitter failed.
The mES plots for data seem to look as expected, i.e., zero contributions
in all bins from flavours 4 and 5 and peaking BG, argus and signal
components look good for flavour 3 (Charged Bs).
There is a problem (also present when the points for mm2 scans look good)
arises from the mES*Vxbn.eps (x=c or u and n=3,4 or5). Looking at the
bins for these mES fits, we have some bins with (without) peaking BG, some
with (without) argus present. Looking at thwe fit results one sees very
large errors for the data points.
Look at the vubfit.out file or see below.
------------------------------------------------------
fitWithError
PARAMETER DEFINITIONS:
NO. NAME VALUE STEP SIZE LIMITS
1 vubincomp 1.00000e+00 1.00000e-03 0.00000e+00 1.00000e+01
2 vcbothcomp 1.00000e+00 1.00000e-03 0.00000e+00 1.00000e+01
3 vuboutcomp 1.00000e+00 1.00000e-03 0.00000e+00 1.00000e+01
4 flag -2.00000e+00 constant
4 flag -3.00000e+00 constant
**********
** 1 **FIX 3
**********
**********
** 2 **FIX 4
**********
PARAMETER 4 IS CONSTANT. IGNORED.
**********
** 3 **MINI 1e+04 0.001
**********
FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.
START MIGRAD MINIMIZATION. STRATEGY 1. CONVERGENCE WHEN EDM .LT.
1.00e-06
FCN=47.4522 FROM MIGRAD STATUS=INITIATE 48 CALLS 49
TOTAL
EDM= unknown STRATEGY= 1 NO ERROR MATRIX
EXT PARAMETER CURRENT GUESS STEP FIRST
NO. NAME VALUE ERROR SIZE DERIVATIVE
1 vubincomp 1.53569e-01 1.00000e-03 -3.95017e-01 -8.75693e+01
2 vcbothcomp 1.34260e-01 1.00000e-03 0.00000e+00 -9.27163e+02
3 vuboutcomp 1.00000e+00 fixed
4 flag -3.00000e+00 constant
MIGRAD MINIMIZATION HAS CONVERGED.
MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.
COVARIANCE MATRIX CALCULATED SUCCESSFULLY
FCN=28.3576 FROM MIGRAD STATUS=CONVERGED 97 CALLS 98
TOTAL
EDM=1.26959e-08 STRATEGY= 1 ERROR MATRIX
ACCURATE
EXT PARAMETER STEP FIRST
NO. NAME VALUE ERROR SIZE DERIVATIVE
1 vubincomp 6.55896e-02 4.30860e-02 1.02883e-04 1.33421e-03
2 vcbothcomp 1.92242e-01 1.46093e-02 2.05107e-05 -9.57320e-03
3 vuboutcomp 1.00000e+00 fixed
4 flag -3.00000e+00 constant
EXTERNAL ERROR MATRIX. NDIM= 25 NPAR= 2 ERR DEF=1
1.858e-03 -4.313e-04
-4.313e-04 2.134e-04
PARAMETER CORRELATION COEFFICIENTS
NO. GLOBAL 1 2
1 0.68489 1.000 -0.685
2 0.68489 -0.685 1.000
**********
** 4 **MINOS 1e+04 0.001
**********
FCN=28.3576 FROM MINOS STATUS=SUCCESSFUL 60 CALLS 158
TOTAL
EDM=1.26959e-08 STRATEGY= 1 ERROR MATRIX
ACCURATE
EXT PARAMETER PARABOLIC MINOS ERRORS
NO. NAME VALUE ERROR NEGATIVE POSITIVE
1 vubincomp 6.55896e-02 4.30860e-02 -4.40159e-02 4.23512e-02
2 vcbothcomp 1.92242e-01 1.46093e-02 -1.43296e-02 1.49194e-02
3 vuboutcomp 1.00000e+00 fixed
4 flag -3.00000e+00 constant
PARAMETER DEFINITIONS:
NO. NAME VALUE STEP SIZE LIMITS
4 flag 2.00000e+00 constant
4 flag 3.00000e+00 constant
**********
** 5 **FIX 3
**********
PARAMETER 3 ALREADY FIXED. IGNORED.
**********
** 6 **FIX 4
**********
PARAMETER 4 IS CONSTANT. IGNORED.
**********
** 7 **MINI 1e+04 0.001
**********
FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.
START MIGRAD MINIMIZATION. STRATEGY 1. CONVERGENCE WHEN EDM .LT.
1.00e-06
FCN=0 FROM MIGRAD STATUS=INITIATE 61 CALLS 220 TOTAL
EDM= unknown STRATEGY= 1 NO ERROR MATRIX
EXT PARAMETER CURRENT GUESS STEP FIRST
NO. NAME VALUE ERROR SIZE DERIVATIVE
1 vubincomp nan 4.30860e-02 0.00000e+00 0.00000e+00
2 vcbothcomp nan 1.46093e-02 0.00000e+00 0.00000e+00
3 vuboutcomp 1.00000e+00 fixed
4 flag 3.00000e+00 constant
MIGRAD MINIMIZATION HAS CONVERGED.
MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.
MINUIT WARNING IN HESSE
============== Second derivative zero for parameter1
MNHESS FAILS AND WILL RETURN DIAGONAL MATRIX.
FCN=0 FROM HESSE STATUS=FAILED 11 CALLS 231 TOTAL
EDM=0 STRATEGY= 1 ERROR MATRIX UNCERTAINTY 100.0
per cent
EXT PARAMETER APPROXIMATE STEP FIRST
NO. NAME VALUE ERROR SIZE DERIVATIVE
1 vubincomp nan nan 0.00000e+00 0.00000e+00
2 vcbothcomp nan nan 0.00000e+00 0.00000e+00
3 vuboutcomp 1.00000e+00 fixed
4 flag 3.00000e+00 constant
MIGRAD MINIMIZATION HAS CONVERGED.
MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.
MINUIT WARNING IN HESSE
============== Second derivative zero for parameter1
MNHESS FAILS AND WILL RETURN DIAGONAL MATRIX.
FCN=0 FROM MIGRAD STATUS=CONVERGED 83 CALLS 242 TOTAL
EDM=0 STRATEGY= 1 ERROR MATRIX UNCERTAINTY 100.0
per cent
EXT PARAMETER APPROXIMATE STEP FIRST
NO. NAME VALUE ERROR SIZE DERIVATIVE
1 vubincomp nan nan 0.00000e+00 0.00000e+00
2 vcbothcomp nan nan 0.00000e+00 0.00000e+00
3 vuboutcomp 1.00000e+00 fixed
4 flag 3.00000e+00 constant
EXTERNAL ERROR MATRIX. NDIM= 25 NPAR= 2 ERR DEF=1
nan nan
nan nan
ERR MATRIX APPROXIMATE
PARAMETER CORRELATION COEFFICIENTS
NO. GLOBAL 1 2
1 0.00000 1.000 0.000
2 0.00000 0.000 1.000
ERR MATRIX APPROXIMATE
**********
** 8 **MINOS 1e+04 0.001
**********
MINUIT WARNING IN HESSE
============== Second derivative zero for parameter1
MNHESS FAILS AND WILL RETURN DIAGONAL MATRIX.
MINUIT WARNING IN MNCROS
============== Cannot find slope of the right sign
POSITIVE MINOS ERROR NOT CALCULATED FOR PARAMETER
1
MINUIT WARNING IN MNCROS
============== Cannot find slope of the right sign
NEGATIVE MINOS ERROR NOT CALCULATED FOR PARAMETER
1
MINUIT WARNING IN MNCROS
============== Cannot find slope of the right sign
POSITIVE MINOS ERROR NOT CALCULATED FOR PARAMETER
2
MINUIT WARNING IN MNCROS
============== Cannot find slope of the right sign
NEGATIVE MINOS ERROR NOT CALCULATED FOR PARAMETER
2
FCN=0 FROM MINOS STATUS=FAILURE 156 CALLS 409 TOTAL
EDM=0 STRATEGY= 1 ERROR MATRIX UNCERTAINTY 100.0
per cent
EXT PARAMETER PARABOLIC MINOS ERRORS
NO. NAME VALUE ERROR NEGATIVE POSITIVE
1 vubincomp nan nan
2 vcbothcomp nan nan
3 vuboutcomp 1.00000e+00 fixed
4 flag 3.00000e+00 constant
Make scan plots
Info in <TCanvas::Print>: eps file
/u/br/clarke1/scra/IbuppR14mm2d2ch-065/ppR14mm2d2ch-065vubin_scan.eps has
been created
Info in <TCanvas::Print>: eps file
/u/br/clarke1/scra/IbuppR14mm2d2ch-065/ppR14mm2d2ch-065vcboth_scan.eps has
been created
Info in <TCanvas::Print>: eps file
/u/br/clarke1/scra/IbuppR14mm2d2ch-065/ppR14mm2d2ch-065vubout_scan.eps has
been created
Doing now the chi2 fit
CHISQ CALCULATION:: 1.35159e+243 1.36621e+161 8.70623e+252
Break of chi2 chop contrib:: 0.291722 0 797.442
Break of chi2 chop contrib:: -1.22235 1 1921.97
Break of chi2 chop contrib:: 1.13429 2 423.731
Break of chi2 chop contrib:: 1.49038 3 94.7063
Break of chi2 chop contrib:: 29.4633 4 0.00161483
Break of chi2 chop contrib:: -1.10929 5 8.17576
Break of chi2 chop contrib:: -1.29868 6 2.98222
Chi Square / NDoF of the Fit :175.218
------------------------------------------------------
C.C.
---------- Forwarded message ----------
Date: Wed, 14 Jun 2006 14:59:32 -0700 (PDT)
From: Chukwudi Kweku Clarke <[log in to unmask]>
To: Heiko Lacker <[log in to unmask]>
Cc: [log in to unmask]
Subject: Re: mm2 scans
Hi Heiko. I look at the point (your 2nd question) you were talking about
and looking at the corresponding results.dat file
(ppR14mm2d2ch-065results.dat) I saw that there is no error for Vub in
fitted 1' bin. This (I think) results in the partial BRBR having on error
also. I will look at the mES fits for this and for your first question.
Cheers, C.C.
On Wed, 14 Jun 2006, Heiko Lacker wrote:
> Hi Chukwudi,
>
> 1. The charged B's for mX look strange.
> Here, it would be worthwhile to understand in more detail if the large
> point-to-point instabilities for larger mm^2 cuts are pointing to mES
> fit problems. Could you have please a closer look?
>
> 2. I have also doubts about the size of the error bars in the diff plots.
> If you compare for example for charged B's in the Pplus analysis for
> selection D the difference between the point at 1 GeV^2 with 0.5 GeV^2
> I would expect an error bar of +-0.0018.
> The error bar plotted is much smaller than that. This looks a bit odd.
>
> Cheers,
> Heiko
>
> On Tue, 13 Jun 2006, Chukwudi Kweku Clarke wrote:
>
> > Hi gang, you may find links to all the mm2 scans on my mm2 webpage;
> > http://www.slac.stanford.edu/~clarke1/protected/clarke1/web/mm2web.html
> > . The names and addresses of the webpages below have changed so please
> > use the above link.
> > Cheers, C.C.
> >
> > On Sun, 11 Jun 2006, CHUKWUDI KWEKU CLARKE wrote:
> >
> > > Hi all, you may find the mm2 scans using binned mES fits below.
> > >
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpage1.html
> > > (mixed Bs 1D analysis).
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpageCh1.html
> > > (Charged Bs 1D analysis).
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpageNu1.html
> > > (Neutral Bs 1D analysis).
> > >
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpage2.html
> > > (mixed Bs 2D analysis).
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpageCh2.html
> > > (Charged Bs 2D analysis).
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpageNu2.html
> > > (Neutral Bs 2D analysis).
> > >
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpage3.html
> > > (mixed Bs Pplus analysis).
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpageCh3.html
> > > (Charged Bs Pplus analysis).
> > > http://www.slac.stanford.edu/~clarke1/protected/clarke1/mywebpageNu3.html
> > > (Neutral Bs Pplus analysis).
> > >
> > > C.C.
> > >
> >
>
