Radiative corrections, quasi-Monte Carlo methods and discrepancy. Computational aspects of high energy phenomenology

In this thesis we looked at the computation of six-fermion processes at LEP2, e.g., the process e-e+→e- anti νeu anti d. A good numerical estimate of these processes is necessary to improve on the measurement of the W-mass. These processes contain Feynman-diagrams with propagating Z and W's, which are massive and have a non-negligible width, i.e., they can decay into other particles. The effect of these widths in cross-sections can be taken into account, but one has to be very careful not to break the gauge invariance of the theory as this would lead in this case to huge errors in the numerical results. (orig./WL)