Parton distributions with LHC data

NNPDF Collaboration, Richard D. Ball, Valerio Bertone, Stefano Carrazza, Christopher S. Deans, Luigi Del Debbio, Stefano Forte, Alberto Guffanti, Nathan P. Hartland, Jose I. Latorre, Juan Rojo, Maria Ubiali

Research output: Contribution to journalArticlepeer-review


We present the first determination of parton distributions of the nucleon at NLO and NNLO based on a global data set which includes LHC data: NNPDF2.3. Our data set includes, besides the deep inelastic, Drell-Yan, gauge boson production and jet data already used in previous global PDF determinations, all the relevant LHC data for which experimental systematic uncertainties are currently available: ATLAS and LHCb W and Z rapidity distributions from the 2010 run, CMS W electron asymmetry data from the 2011 run, and ATLAS inclusive jet cross-sections from the 2010 run. We introduce an improved implementation of the FastKernel method which allows us to fit to this extended data set, and also to adopt a more effective minimization methodology. We present the NNPDF2.3 PDF sets, and compare them to the NNPDF2.1 sets to assess the impact of the LHC data. We find that all the LHC data are broadly consistent with each other and with all the older data sets included in the fit. We present predictions for various standard candle cross-sections. and compare them to those obtained previously using NNPDF2.1. and specifically discuss the impact of ATLAS electroweak data on the determination of the strangeness fraction of the proton. We also present collider PDF sets, constructed using only data from HERA. the Tevatron and the LHC. but find that this data set is neither precise nor complete enough for a competitive PDF determination.

Original languageEnglish
Pages (from-to)244-289
Number of pages46
JournalNuclear physics b
Issue number2
Publication statusPublished - 11 Feb 2013

Cite this