Table of Contents
Fetching ...

NNLO global parton analysis

A. D. Martin, R. G. Roberts, W. J. Stirling, R. S. Thorne

TL;DR

The paper performs NNLO and LO global analyses of DIS and related hard scattering data to extract parton distributions, using recent HERA DIS data and Tevatron jet data. Since exact NNLO splitting functions were not fully known, it uses constrained NNLO splittings from van Neerven and Vogt and higher moments to bound evolution, with slow and fast extremes. Results show modest global improvement at NNLO over NLO; parton distributions shift by up to about 10% at $Q^2=10\,\mathrm{GeV}^2$, especially for $x<0.01$, and the high-x gluon is adjusted to accommodate Tevatron jets. Notably, LO partons fit Tevatron jet data surprisingly well, while NNLO slightly worsens the jet fit; predictions for $F_L$ and vector boson production remain physically sensible and uncertainties shrink with improved splitting-function constraints.

Abstract

We perform a NNLO (and a LO) global parton analysis in which we include the new precise data for deep inelastic scattering from HERA and for inclusive jet production at the Tevatron, together with the improved knowledge of the three-loop splitting functions. The results are compared with our recent NLO analyses. The LO fit produces significantly worse results in general, but gives a surprisingly good fit to the Tevatron high-E_T jet data. For the approximate NNLO analysis we notice a slight improvement in the quality of the global fit, and find that the partons are changed by up to 10% at Q^2=10 GeV^2$, in particular in the x<0.01 regime.

NNLO global parton analysis

TL;DR

The paper performs NNLO and LO global analyses of DIS and related hard scattering data to extract parton distributions, using recent HERA DIS data and Tevatron jet data. Since exact NNLO splitting functions were not fully known, it uses constrained NNLO splittings from van Neerven and Vogt and higher moments to bound evolution, with slow and fast extremes. Results show modest global improvement at NNLO over NLO; parton distributions shift by up to about 10% at , especially for , and the high-x gluon is adjusted to accommodate Tevatron jets. Notably, LO partons fit Tevatron jet data surprisingly well, while NNLO slightly worsens the jet fit; predictions for and vector boson production remain physically sensible and uncertainties shrink with improved splitting-function constraints.

Abstract

We perform a NNLO (and a LO) global parton analysis in which we include the new precise data for deep inelastic scattering from HERA and for inclusive jet production at the Tevatron, together with the improved knowledge of the three-loop splitting functions. The results are compared with our recent NLO analyses. The LO fit produces significantly worse results in general, but gives a surprisingly good fit to the Tevatron high-E_T jet data. For the approximate NNLO analysis we notice a slight improvement in the quality of the global fit, and find that the partons are changed by up to 10% at Q^2=10 GeV^2$, in particular in the x<0.01 regime.

Paper Structure

This paper contains 1 section, 5 equations, 6 figures, 3 tables.

Table of Contents

  1. Acknowledgments

Figures (6)

  • Figure 1: The description of data for the $F_2$ structure function at a few representative values of $x$ obtained in the LO, NLO MRST2001 and NNLO global analyses.
  • Figure 2: The comparison of the quality of the LO MRST and the NLO MRST fit to the CDF1B high-$E_T$ jet data CDF. The open points are the NLO values before the correlated systematic errors have been considered while the solid points are those for the LO fit.
  • Figure 3: Gluon distributions obtained from the LO, NLO and NNLO analyses at various values of $Q^2$. The three NNLO gluons result from the slow and fast extremes of the splitting functions together with the average of these.
  • Figure 4: A comparison of the gluon, up and down distributions obtained in the NNLO analysis with those obtained in the NLO fit MRST2001. The comparison is shown for two values of $Q^2$.
  • Figure 5: Predictions for $F_L$ from the LO, NLO and NNLO partons.
  • ...and 1 more figures