An analytic parton shower. Algorithms, implementation and validation

The realistic simulation of particle collisions is an indispensable tool to interpret the data measured at high-energy colliders, for example the now running Large Hadron Collider at CERN. These collisions at these colliders are usually simulated in the form of exclusive events. This thesis focuses on the perturbative QCD part involved in the simulation of these events, particularly parton showers and the consistent combination of parton showers and matrix elements. We present an existing parton shower algorithm for emissions off final state partons along with some major improvements. Moreover, we present a new parton shower algorithm for emissions off incoming partons. The aim of these particular algorithms, called analytic parton shower algorithms, is to be able to calculate the probabilities for branchings and for whole events after the event has been generated. This allows a reweighting procedure to be applied after the events have been simulated. We show a detailed description of the algorithms, their implementation and the interfaces to the event generator WHIZARD. Moreover we discuss the implementation of a MLM-type matching procedure and an interface to the shower and hadronization routines from PYTHIA. Finally, we compare several predictions by our implementation to experimental measurements at LEP, Tevatron and LHC, as well as to predictions obtained using PYTHIA. (orig.)