The CONF draft 1.2 is prepared and awaiting sign-off from EB chair
ttW results will be shown at Moriond EW on Friday!
ttHML analysis status:
Given the updated timescale for the analysis (EB request by end of May), we need to freeze the MVAs strategy by Mid-End April. First step of optimization concerns the list of final variables to be used. We are considering to start by dropping correlated variables if it does not have a significant effect on the performance however the method does not seem to be the optimal way. Suggestions to use a more systematic approach (backward feature elimination) where e.g last ranked variables are dropped one by one (or by group) is proposed but we need to see if it is feasible given the time it takes to train.