Status of QEL Analysis

1
Status of QEL Analysis

• QEL-like Event Selection and Sample
• ND Flux Extraction
• Fitting for

Mark Dorman UCL / RAL
MINOS Collaboration Meeting FNAL, 7th-10th
December '06
2
QEL-like Event Selection W (or ) Cut
Apply cut at Wlt1.3

• Large data excesses compared to MC
• and slight shift to higher in data.
• Cedar and Daikon are an improvement
• here but cannot resolve the differences.

3
QEL-like Event Selection 0

• Much more pure QEL sample than
• including events with small showers.
• Shape improvements with Daikon for
• muon angle distribution.

4
PDF-Based Selection with Cedar and Daikon

• Make PID for
• different MC settings.
• Select events from
• independent sample
• with each of these
• PIDs to compare.

5
Flux Extraction Methodology

• come from NEUGEN.
• are the selection
  efficiencies estimated with MC.
• 
  is the number of NC events at
• neutrino energy E where refers to the
  MC flux.
• There is a partial energy unfolding as the
  equation is utilized in bins
• of reconstructed energy but the denominators
  in the efficiencies are
• are equal to the numbers of events generated
  in true energy bins.
• It is impossible to remove the MC flux from
  this equation and so the
• next step is to iterate until some convergence
  criterion for is met.

6
(No Transcript)
7
and the QEL Cross Section

• A, B and C depend on (amongst other things)
  the axial vector form
• factor of the nucleon which can be written as
• The quasi-elastic axial vector mass can be
  treated as the one free
• parameter of this differential cross section
  and so the QEL-like
• sample distribution can be used to fit
  this constant.

known from neutron decay experiments
MINOS nominal value is 1.032, K2K recently fitted
to get 1.2
8
'Fitting' for

• I have reweighted a MC sample for various
  different values
• I then take an independent sample weighted to a
  different value
• and use the ROOT - test to map out the
  v.s. space.
• I then fit the minimum with a quadratic to
  extract a best fit .

Wlt1.3 Selection
9
'Fitting' for

• Mock data set Cedar/Carrot, 1.057
• Fit returns

From looking for ? 1

• Mock data set Cedar/Daikon,
• Fit returns
• Slightly low, expected(?) with updated
• form factors ( other changes) in Daikon.

10
'Fitting' for

• This is only a preliminary step and there are
  many things to consider
• for further work such as
• The main goal here is to try to combine fits
  for cross section parameters
• that affect the shape and normalization of the
  QEL-like sample into the
• iterative flux extraction.
• For flux extraction need to know
  ,
• Also need to factor in INTRANUKE reweighting...

• Fitting distributions in bins of and
  comparing/combining results.
• Which QEL sample selection to use (statistics
  v.s. purity v.s. ...).
• Adding other cross section parameters to affect
  the non-QEL events
• such as or the KNO parameters.
• More rigorous approach constructing event
  predictions based on cross
• section formula, flux, sample
  efficiencies/purities, conversion from

11
Summary

• Data/MC QEL-like sample discrepancies remain in
  Cedar/Daikon.
• So far detector/instrumentation effects
  have been ruled out as a
• possible solution still some to do
  however...
• Starting to play with cross-section parameter
  fitting for the QEL-like
• sample although this is only very preliminary.
• Will re-extract ND flux using Cedar/Daikon as
  statistics increase.
• Need to implement iterative flux extraction
  and possibly incorporate
• fitting for cross-section parameters into this
  method.
• Different INTRANUKE samples were interesting
  for Birch/Carrot and
• had a big effect on the QEL-like sample and
  extracted flux. Will redo
• studies in Cedar/Daikon with new reweighting
  scheme.