QCD processes and search for supersymmetry at the LHC

In this thesis, a data-driven method to estimate the number of QCD background events in a multijet search for supersymmetry at the LHC was developed. The method makes use of two models which predict the correlation of two key search variables, the missing transverse momentum and an angular variable, in order to extrapolate from a QCD dominated control region to the signal region. A good performance of the method was demonstrated by its application to 36 pb􀀀1 data, taken by the CMS experiment in 2010, and by the comparison with an alternative method. Comparing the number of data events to a combined background expectation of QCD and data-driven estimates of the electroweak and top background, no statistically significant excess was observed for three pre-defined search regions. Limits were calculated for the (m0,m1/2) parameter space of the cMSSM, exceeding previous measurements. The expected sensitivity for further refined search regions was investigated.