Matching NLO QCD computations and parton shower simulations
S. Frixione, B. R. Webber
TL;DR
The paper develops MC@NLO, a framework to marry NLO QCD calculations with parton-shower Monte Carlo simulations to produce fully exclusive, hadronized events with NLO-accurate total rates. It introduces a modified subtraction strategy to avoid double counting and large weights, ensuring correct NLO expansion while preserving the shower's soft/collinear resummation. A toy model clarifies the method and its key ingredients, and the approach is demonstrated through observables that expose both fixed-order accuracy and shower-like logarithmic behavior. The work enables more precise and exclusive predictions for collider phenomenology, bridging fixed-order and resummed descriptions and facilitating interfacing with hadronization models.
Abstract
We propose a method for matching the next-to-leading order (NLO) calculation of a given QCD process with a parton shower Monte Carlo (MC) simulation. The method has the following features: fully exclusive events are generated, with hadronization according to the MC model; total exclusive rates are accurate to NLO; NLO results for distributions are recovered upon expansion in $α_S$; hard emissions are treated as in NLO computations while soft/collinear emissions are handled by the MC simulation, with the same logarithmic accuracy as the MC; and matching between the hard- and soft/collinear-emission regions is smooth. A fraction of events with negative weight is generated, but unweighting remains possible with reasonable efficiency. The method is clarified using a simple toy model, and illustrated by application to the hadroproduction of W$^+$W$^-$ pairs.