1 Summary and Conclusions Kyle Cranmer (New York University) Harrison B. Prosper (Florida State University) Sezen Sekmen (CERN) LPCC Workshop: Likelihoods for LHC Searches LPCC Workshop on Likelihoods CERN
List of Talks Day 1 SezenGoals GlenPrinciples KyleContext/Scope Feedback Marco Maggie Béranger Day 2 KyleHistFactory SvenATLAS HZZ4l Higgs Combination MinshuiCMS HaoshuangATLAS Day 3 Wolfgang Javier (thanks Maurizio!) Wouter Panelists Sünje, Mike, Lorenzo 2LPCC Workshop on Likelihoods CERN
DAY 1 LPCC Workshop on Likelihoods CERN3
Sezen: Workshop Goals Goals Educate ourselves: why are likelihoods needed? Move towards routine publication of likelihoods LPCC Workshop on Likelihoods CERN4
Glen: Basic Ideas Distribution Probability density (or mass) function, Nature(x) xpotential observations Model P(x | μ, θ) is a parametric model of the unknown function Nature(x) with parameters μ and θ, some of which are interesting (μ) and some not (θ). Likelihood L(μ, θ) = L(D | μ, θ) = P(D | μ, θ) D = observed data LPCC Workshop on Likelihoods CERN5
Glen: Basic Ideas Need a way to get rid of parameters not of current interest. There are two general ways, marginalization and profiling: Marginal Likelihood Profile Likelihood Profiling can be regarded as marginalization with the prior LPCC Workshop on Likelihoods CERN6
Kyle: Context & Scope LPCC Workshop on Likelihoods CERN7
Feedback
Marco: Is it the SM Higgs? LHC Higgs Cross Section Working Group Assumptions SM tensor structure (CP-even scalar) A single zero-width resonance κ i = σ i / σ SMi and κ f = Γ f / Γ SMi are free parameters, where How do we best report experimental results (with the goal of allowing more detailed/accurate studies)? LPCC Workshop on Likelihoods CERN9
Maggie: Is it the SM Higgs? LPCC Workshop on Likelihoods CERN10 Can use an effective field theory (EFT) approach:
Maggie: Is it the SM Higgs? LPCC Workshop on Likelihoods CERN11
Béranger: Is it the SM Higgs? Effective Lagrangian Fitting procedure LPCC Workshop on Likelihoods CERN12
Béranger: Is it the SM Higgs? LPCC Workshop on Likelihoods CERN13
DAY 2 LPCC Workshop on Likelihoods CERN14
Kyle: HistFactory Equivalent to a multi-bin Poisson model with bins so small that the chance of > a single count per bin is negligible n is the number of events and {x e } are the measurements (e.g., the di-photon masses) In general, f is a mixture: LPCC Workshop on Likelihoods CERN15
Kyle: HistFactory which, in this case, represents a Gaussian G(x| μ, σ). f p (a p | α p ) are the likelihoods of the auxiliary measurements a p from either real, simulated, or hypothetical experiments. These functions provide constraints on the parameters α and hence on the parameters ν c (α). LPCC Workshop on Likelihoods CERN16
Kyle: HistFactory LPCC Workshop on Likelihoods CERN17 XML representation of model Kyle HistFactory RooWorkspace
Sven: HZZ*(4l) in ATLAS LPCC Workshop on Likelihoods CERN18
Sven: HZZ*(4l) in ATLAS LPCC Workshop on Likelihoods CERN19 Cranmer, K, Kernel Estimation in High-Energy Physics Computer Physics Communications 136: , 2001 hep ex/ Kernel density estimation + density morphing + HistFactory
Sven: HZZ*(4l) in ATLAS LPCC Workshop on Likelihoods CERN20 Editorial comment: Jack’s intuition is spot on! For discrepant results, the combined result ought to be worse.
Sven: HZZ*(4l) in ATLAS Clarity Prize goes to Sven for explaining to me why a p-value computed from the background-only hypothesis depends on the alternative hypothesis! Harrison: “Please explain this plot” Sven: “The sampling distribution of t(x) = -2 ln L p /L max is independent of m H, as it should be, but the power of the test is maximized for each m H, so the observed value of t changes with m H ” LPCC Workshop on Likelihoods CERN21
Higgs Combination
Mingshui: Higgs Combination (CMS) Model: Marked Poisson Process (see Kyle’s HistFactory talk) LEP No constraints for parameters θ with systematic uncertainties Tevatron Use priors π(θ|θ 0 ) to constrain θ LHC Interpret π(θ|θ 0 ) as π(θ|θ 0 ) ~ f (θ 0 |θ) π(θ) Cowanscher Ur-prior! and interpret f (θ 0 |θ) as the likelihood for auxiliary measurements θ 0 LPCC Workshop on Likelihoods CERN23
Mingshui: Higgs Combination (CMS) Assumptions (current measurements) Data are disjoint Standard Model with m H and μ as free parameters Same m H for all channels Detailed models can be provided in RooWorkspace form LPCC Workshop on Likelihoods CERN24
Haoshuang: Higgs Combination (ATLAS) Basic tool is HistFactory for all channels except for H to γγ A Single Channel LPCC Workshop on Likelihoods CERN25
Haoshuang: Higgs Combination (ATLAS) Important point In combining channels the Greek symbol fallacy is avoided. An explicit decision must be made about how parameters with the same name are related, if at all. Typically done by modifying the XML representation of the model. LPCC Workshop on Likelihoods CERN26
DAY 3 LPCC Workshop on Likelihoods CERN27
Wolfgang: BSM Searches Guided by a well-motivated theory, e.g., the pMSSM, and its simplified model decomposition pMSSM Results (non-CMS) …but CMS pMSSM / SMs analysis in progress… LPCC Workshop on Likelihoods CERN28
Wolfgang: BSM Searches LPCC Workshop on Likelihoods CERN29
Javier: BSM Searches LPCC Workshop on Likelihoods CERN30
Javier: BSM Searches LPCC Workshop on Likelihoods CERN31
Javier: BSM Searches LPCC Workshop on Likelihoods CERN32
Javier: BSM Searches LPCC Workshop on Likelihoods CERN33 Nuisance parameters marginalized through Monte Carlo integration
Wouter: RooFit RooFit is a probability modeling language: RooStats provides high level statistical tools that use RooFit models LPCC Workshop on Likelihoods CERN34
Wouter: RooFit A RooWorkspace is a mechanism to store a model + data LPCC Workshop on Likelihoods CERN35
Panel Discussion Sünje, Mike, Lorenzo HEPData on INSPIRE Make data sets searchable, findable, citable Assign Digital Object Identifier (DOI) to data Should we track the re-use of data? Should we have a single portal (e.g, Inspire)? Will will have a single portal? Will need non-web access also RECAST requests that are honored could yield citation Are there legal issues? LPCC Workshop on Likelihoods CERN36
CONCLUSIONS LPCC Workshop on Likelihoods CERN37
ICHEP 2040 LPCC Workshop on Likelihoods CERN38 Data pNMSSM OTTRTA SM m e, m μ, m τ m u, m d, m s, m c, m b, m t θ 12, θ 23, θ 13, δ g 1, g 2, g 3 θ QCD μ, λ The New Standard Model has been firmly established
Conclusions We could do a better job of understanding the LHC data if more information were made public in a systematic way A general way to do this is to publish the probability model + relevant data set The technology exists (RooWorkspace, Inspire, HepData) to publish arbitrarily complicated models, retrieve them and use them in analyses My sense is that our field is nearing a tipping point, for the better! LPCC Workshop on Likelihoods CERN39
Thanks! We thank the LHC Physics Centre at CERN (LPCC) for hosting this workshop and its financial support of two RooStats developers. We thank the Theory Secretariat for organizing the coffee breaks! We thank YOU for making this workshop both informative and enjoyable. We thank the World’s funding agencies and the World’s taxpayers for their generous support: LHC cost: $1million / scientist LPCC Workshop on Likelihoods CERN40