Statistical methods in LHC data analysis

Europe/Zurich
CERN

CERN

Luca Lista (INFN - Naples)
Description
Those lectures are intended as the first part of a wider tutorial aiming at the use of RooFit and RooStats. The purpose is to present selected statistics topics used for LHC data analysis manageable with RooFit and RooStats toolkits that will be presented on November 26 and 27 2009 EVO: Statistical methods (Universe)
  • Wednesday, 18 November
    • 09:30 09:40
      Introduction 10m 4-3-006 TH Theory Conference Room

      4-3-006 TH Theory Conference Room

      CERN

      Speaker: Luca Lista (INFN - Naples)
      Slides
    • 09:40 10:55
      Part I.1 1h 15m 4-3-006 TH Theory Conference Room

      4-3-006 TH Theory Conference Room

      CERN

      Parameter estimates Likelihood function Maximum Likelihood method Problems with asymmetric errors
      Speaker: Luca Lista (INFN - Naples)
      How to playback the recording
      Slides
      Video recording
    • 10:55 11:00
      break 5m
    • 11:00 12:20
      Part I.2 1h 20m 4-3-006 TH Theory Conference Room

      4-3-006 TH Theory Conference Room

      CERN

      Hypothesis testing Neyman-Pearson lemma and likelihood ratio Multivariate analysis (elements) Confidence intervals Feldman-Cousins ordering Chi-squared fits and goodness-of-fit
      Speaker: Luca Lista (INFN - Naples)
      How to play the video
      Slides
      video recording
  • Thursday, 19 November
    • 09:30 10:50
      Part II.1 1h 20m 40-S2-A01

      40-S2-A01

      CERN

      Bayes theorem Bayesian probability Bayesian inference
      Speaker: Luca Lista (INFN - Naples)
      Slides
      Video
    • 10:50 11:00
      break 10m
    • 11:00 12:25
      Part II.2 1h 25m 40-S2-A01

      40-S2-A01

      CERN

      Upper limits Treatment of background Bayesian limits Modified frequentist approach (CLs method) Nuisance parameters and Cousins-Highland approach
      Speaker: Luca Lista (INFN - Naples)
      Slides
      Video