Towards parton distribution functions with theoretical uncertainties

TL;DR

This paper introduces a new method to incorporate theoretical uncertainties, especially missing higher order uncertainties, into parton distribution function fits using scale variation techniques.

Contribution

It proposes two novel approaches for estimating and integrating theoretical uncertainties into PDF fits, enhancing their accuracy and reliability.

Findings

Estimation of MHOUs via scale variation improves PDF uncertainty quantification.

Construction of a theoretical covariance matrix allows combined experimental and theoretical uncertainty analysis.

Methods provide a systematic way to include theoretical errors in PDF determinations.

Abstract

An important limitation in current fits of parton distribution functions (PDFs) is that PDF uncertainties do not include any source of theoretical uncertainty. Here we present a general method for incorporating theoretical uncertainties into PDF fits, focussing in particular on perturbative missing higher order uncertainties (MHOUs). We consider two methods for estimating the effect of MHOUs on PDFs, both based on scale variations. Firstly, we present PDF fits based on theoretical predictions with varied scales, and use these to estimate the associated MHOUs. Secondly, we discuss the construction of a theoretical covariance matrix using scale variations, and its combination with the experimental covariance matrix currently used in PDF fits.