A new generation of parton distribution functions with increased precision and quantitative estimates of uncertainties is presented. This work significantly extends previous CTEQ and other global analyses on two fronts: (i) a full treatment of available experimental correlated systematic errors for both new and old data sets; (ii) a systematic and pragmatic treatment of uncertainties of the parton distributions and their physical predictions, using a recently developed eigenvector-basis approach to the Hessian method. The new gluon distribution is considerably harder than that of previous standard fits. A number of physics issues, particularly relating to the behavior of the gluon distribution, are addressed in more quantitative terms than before. Extensive results on the uncertainties of parton distributions at various scales, and on parton luminosity functions at the Tevatron RunII and the LHC, are presented. The latter provide the means to quickly estimate the uncertainties of a wide range of physical processes at these high-energy hadron colliders, based on current knowledge of the parton distributions. In particular, the uncertainties on the production cross sections of the W, Z at the Tevatron and the LHC are estimated to be ±4% and ±5% respectively, and that of a light Higgs at the LHC to be ±5%.Progress on the determination of the parton distribution functions (PDF's) of the nucleon, from global quantum chromodynamics (QCD) analysis of hard scattering processes, is central to precision standard model (SM) phenomenology, as well as to new physics searches, at lepton-hadron and hadron-hadron colliders. There have been many new developments in recent years, beyond the conventional analyses that underlie the widely used PDF's [1-3]. These developments have been driven by the need to quantify the uncertainties of the PDF's and their physical predictions [4][5][6][7][8][9][10][11][12]. We report in this paper on a comprehensive new global QCD analysis based on the most current data, and on the recently developed methods of uncertainty study of [10][11][12]. This new analysis includes a full treatment of all available correlated experimental errors, as well as an extensive exploration of the parametrization of the input nonperturbative PDF's.Although this work is built on the series of previous CTEQ parton distributions [2], it represents more than an evolutionary updating of previous work to incorporate new experimental data sets. The methodology of [10-12] goes beyond the traditional paradigm of producing some subjectively chosen "best fits." It introduces a set of efficient and practical tools to characterize the parton parameter space in the neighborhood of the global minimum. This allows the systematic exploration of the uncertainties of parton distributions and their physical predictions due to known experimental errors and due to the input theoretical model parameters.There are many complex issues involved in a comprehensive global parton distribution analysis. Foremost among these on the experimental sid...

An up-to-date global QCD analysis of high energy lepton-hadron and hadronhadron interactions is performed to better determine the gluon and quark parton distributions in the nucleon. Improved experimental data on inclusive jet production, in conjunction with precise deep inelastic scattering data, place good constraints on the gluon over a wide range of x; while new data on asymmetries in Drell-Yan processes contribute to better determine the d/u ratio. Comparisons with results of other recent global analyses are made, and the differences are described. Open issues and the general problem of determining the uncertainties of parton distributions are discussed.

The impact of recent precision measurements of DIS structure functions and inclusive jet production at the Fermilab Tevatron on the global QCD analysis of parton distribution functions is studied in detail. Particular emphasis is placed on exploring the range of variation of the gluon distribution G(x,Q) allowed by these new data. The strong coupling of G(x,Q) with ␣ s is fully taken into account. A new generation of CTEQ parton distributions, CTEQ4, is presented. It consists of the three standard sets ͓modified minimal subtraction (MS), deep inelastic scattering ͑DIS͒, and leading order ͑LO͔͒, a series that gives a range of parton distributions with corresponding ␣ s 's, and a set with a low starting value of Q. Previously obtained gluon distributions that are consistent with the high E t jet cross section are also discussed in the context of this new global analysis.

Jet production at the Tevatron probes some of the smallest distance scales currently accessible.A gluon distribution that is enhanced at large x compared to previous determinations provides a better description of the Run 1b jet data from both CDF and DØ. However, considerable uncertainty still remains regarding the gluon distribution at high x. In this paper, we examine the effects of this uncertainty, and of the remaining uncertainties in the NLO QCD theory, on jet cross section comparisons to Run 1b data. We also calculate the range of contributions still possible from any new physics. Predictions are also made for the expanded kinematic range expected for the ongoing Run 2 at the Tevatron and for the LHC. I. INTRODUCTION: JET CROSS SECTIONS AT THE TEVATRONJet production at the Tevatron probes the highest momentum transfer region currently accessible. As this region is potentially sensitive to a wide variety of new physics, there was great interest when the inclusive jet cross section measured by the CDF collaboration in Run 1b exhibited an excess in the highest E T range, when compared to NLO predictions using then-current parton distribution functions [1]. In an attempt to determine if conventional physics could explain the deviation, the CTEQ PDF fitting group carried out a global analysis using information from deep-inelastic scattering and Drell-Yan data, as well as from jet data from the Tevatron, but giving a higher statistical emphasis to the high E T jet data from CDF [2]. NLO predictions using the resulting fit (CTEQ4HJ) reduced the size of the excess observed by CDF. The jet data from both CDF and DØ were also used in a more conventional fit (CTEQ4M) where no special emphasis was given to the high E T data [3].The dominant subprocess that contributes to jet production at high E T is quark-(anti)quark scattering. However, the quark distributions in the corresponding x range are very well constrained by the precise DIS and DY data used in the global fits. Only the gluon distribution has the flexibility to change significantly in the high x region, and indeed the gluon distribution increases by roughly a factor of 2 for x values of about 0.5 in the CTEQ4HJ fit. The gluon-quark scattering subprocesses increase from approximately 20% of the total jet cross section at high E T using CTEQ4M to 40% using CTEQ4HJ [2]. The next group of PDF's from CTEQ (CTEQ5) [4] also contained two sets: CTEQ5M, the standard lowest χ 2 solution, and CTEQ5HJ, defined by a similar statistical enhancement applied to the high E T data from CDF [5] and DØ [6].More recently, DØ has measured the inclusive jet cross section as a function of rapidity y over the range 0 ≤ |y| ≤ 3 [7]. This data set, comprised of 90 data points, has a greater statistical power in the global fits than the CDF and DØ jet cross section measurements in the central region alone. The CTEQ6M fit [8] utilizes these DØ jet cross section measurements, along with the CDF measurements in the central rapidity region, as well as the most recent DIS data from HERA an...

Existing calculations of heavy quark production in charged-current and neutral current lepton-hadron scattering are formulated differently because of the artificial distinction of "light" and "heavy" quarks made in the traditional approach. A proper QCD formalism valid for a wide kinematic range from near threshold to energies much higher then the quark mass should treat these processes in a uniform way. We formulate a unified approach to both types of leptoproduction processes based on the conventional factorization theorem. In this paper, we present the general framework with complete kinematics appropriate for arbitrary masses, emphasizing the simplifications provided by the helicity formalism. We illustrate this approach with an explicit calculation of the leading order contribution to the quark structure functions with general masses. This provides the basis for a complete QCD analysis of charged current and neutral current leptoproduction of charm and bottom quarks to be presented in subsequent papers.

A new implementation of the general PQCD formalism of Collins, including heavy quark mass effects, is described. Important features that contribute to the accuracy and efficiency of the calculation of both neutral current (NC) and charged current (CC) processess are explicitly discussed. This new implementation is applied to the global analysis of the full HERA I data sets on NC and CC cross sections, with correlated systematic errors, in conjunction with the usual fixed-target and hadron collider data sets. By using a variety of parametrizations to explore the parton parameter space, robust new parton distribution function (PDF) sets (CTEQ6.5) are obtained. The new quark distributions are consistently higher in the region x ∼ 10 −3 than previous ones, with important implications on hadron collider phenomenology, especially at the LHC. The uncertainties of the parton distributions are reassessed and are compared to the previous ones. A new set of CTEQ6.5 eigenvector PDFs that encapsulates these uncertainties is also presented. *

No abstract

We develop a general method to quantify the uncertainties of parton distribution functions and their physical predictions, with emphasis on incorporating all relevant experimental constraints. The method uses the Hessian formalism to study an effective chi-squared function that quantifies the fit between theory and experiment. Key ingredients are a recently developed iterative procedure to calculate the Hessian matrix in the difficult global analysis environment, and the use of parameters defined as components along appropriately normalized eigenvectors. The result is a set of 2D eigenvector basis parton distributions ͑where dϷ16 is the number of parton parameters͒ from which the uncertainty on any physical quantity due to the uncertainty in parton distributions can be calculated. We illustrate the method by applying it to calculate uncertainties of gluon and quark distribution functions, W boson rapidity distributions, and the correlation between W and Z production cross sections.