A new approach to matrix element re-weighting

(eng) Hadron collider, like the Tevatron at Fermilab and the Large hadron Collider (LHC) at CERN require some advances analysis technique in order to measure precisely the theoretical parameter. Recently D0 and CDF use the Matrix Element Method, method based on a matrix element re-weighting in order to achieve to measure the top-quark mass with a precision of approximatively one GeV. This re-weighting consist in the convolution of the square matrix element with a transfer functions which parameterize the answer of the detector. Recently the importance of such re-weighting rise in importance and was uses in many other analyses. This thesis presents different method in order the use the matrix element re-weighting. In order to perform efficiently such type of re-weighting a dedicated phase-space integrators was designed (MadWeight). Indeed the numerical evaluation of such convolution requires some advances knowledge in Monte-Carlo techniques. This thesis presents how MadWeight finds an efficient way to re-weight any type of decay chain associate to arbirary transfer functions. Finally, the impact of Initial State Radiation have been studied for the first time and several stategies have been proposed in order to correct those effects.