Skip to content
ShujieL edited this page Nov 9, 2023 · 4 revisions

Welcome to the CJ-database wiki! For now all info in one page - will make it better later

Dataframe format

(Data are provided in both xlsx and csv format under directory data/dataframe, can be easily read into a dataframe)

Headers - generalities

In general:

  • Extracted data: prepend a * (star) sign if the column was extracted using info from original source

Measurements:

The measured values are given by:

  • value: the value of the measurement
  • units: the units (e.g. mb, fb, ....)
    • If dimensionless, use 1

Error headers:

  • uncorrelated errors: append _u to whatever name you like.
    • Example: stat_u is the uncorrelated statistical error; syst_u the uncorrelated systematic error
  • correlated errors: append _c to whatever name you like.
    • Example: norm_c is the overall normalization error.
  • relative vs. absolute errors: Prepend a % sign if error expressed as percent of the quoted value of the observable; otherwise, error interpreted as absolute value.

Process characterization

DIS

A complete characterization of an unpolarized DIS cross section requires specification of the following information

  • obs: the kind of observable (see below)
  • target: the kind of target (proton, deuteron, ratios, heavier nuclei)
  • lepton beam: the flavor and charge of the lepton beam
  • current: the vector boson current (neutral or charged)

Kinematic variables

  • X or x: Bjorken x
  • Q2 or q2: four momentum squared (GeV^2/c^2)
  • Y or y: invariant inelasticity
  • Elab: Lab energy of the collision (GeV/c)

Allowed values for each column

obs

We follow the PDG conventions for structure functions.

  • F2: F2 structure function
  • FL: longitudinal FL
  • R: for the longitudinal to transverse cross section ratio
  • sig_r: reduced DIS cross section (remove the coefficient in front of F2)

Note that ratios of structure functions or cross sections are indicated in the target column. E.g., a ratio of proton to deuteron cross sections woudl be indicated as p/d in that column. Here only the original cross section is indicated.

target:

  • p for protons
  • d for per-nucleon deuteron target observables, i.e, 0.5*obs(d)
    • Note: We may want however to use N for "nucleons", instead
  • p/d for the proton/deuteron ratio, e.g., from NMC
  • n/d for the neutron/deuteron ratio, e.g., from BONUS
  • Nuclei can also be included in the future, e.g.,
    • A_12_6 for 12C
    • A_207_82 for 207Pb
    • attention: the Dueteron would be A_2_1 --> so that in the current conventions A_2_1 = 2*d

lepton beam:

Combine a flavor and charge labels:

  • flavors: e, mu, tau, nu_e, nubar_e, ...
  • charges: plus, minus, also nothing if unknown

Examples:

  • e_plus for e+ beam
  • mu_minus for mu- beams
  • mu if no charge is specified in the original paper (it means Q2 is low enough for gamma-Z interference to be negligible)

current

  • NC = neutral current
  • CC = charged current

Plain text format (deprecated)

The plain data files in the data/text directory are used as input files to the CJ global fitting program.

Filename convention:

{experiment name}_{observable}{target}

*An additional _c as suffix indicates the correlated errors are provided.

Data file example:

In data/text/slac_p_c:

slac,F2,p target ------------- data content

converted from 10010.xlsx ----- additional notes (in this case, the data table ID)

*

0.0210 13 ------------------- normalization factor in fraction, and number of correlated errors

In the header column: x Q2 F2 Ebeam stat sys_tot sys_u dsy dse drc ...

  • stat: total statistical error
  • sys_tot: total systematic error (all correlated and uncorrelated errors added in quadrature)
  • day, dse, doc...: names of the 13 sources of correlated error.