Date
Journal Title
Journal ISSN
Volume Title
Publisher
The analysis presented in this thesis expects to provide the first observation of the opposite sign WW with an associated photon process and to measure its fiducial cross section. The analysis uses proton-proton collision data collected between 2015 and 2018 using the ATLAS detector at a center of mass energy of 13 TeV corresponding to an integrated luminosity of 139 inverse fb. Only events where one W boson decays to an electron and the other W boson decays to a muon are considered. Significant deviations from the measured opposite sign WW with an associated photon production cross section and the Standard Model (SM) prediction can provide evidence for beyond the Standard Model (BSM) physics. In addition, any deviation from the SM prediction can be parameterized in the framework of an effective field theory (EFT) and limits can be placed on these EFT parameters. The results of the EFT measurement are beyond the scope of this thesis and will be presented in another thesis. The majority of the background events arise from processes with prompt photons including tt with an associated photon and Z boson production with an associated photon. The analysis also considers backgrounds from non-prompt photons from electrons or hadrons. To improve the measurement sensitivity, a BDT (boosted decision tree) is trained using several well modeled input variables. The BDT combines the information from the input variables into a single variable: the BDT score. The opposite sign WW with an associated photon fiducial cross section and statistical significance are calculated by performing a binned maximum likelihood fit using the BDT output distribution. The expected statistical significance of the measurement is 7.2 σ and the expected fiducial cross section is 10.5 ± 14% (cross section) ± 15% fb (measurement). Where the cross section uncertainty is associated with the determination of the fiducial volume and the measurement uncertainty is associated with the determination of the signal strength using the binned maximum likelihood fit.