Durham1 Collection of algorithms in ROOT Advanced Multivariate & Statistical Techniques FermiLab Workshop May 31, 2002 Ren é Brun CERN.

Slides:



Advertisements
Similar presentations
S.Towers TerraFerMA TerraFerMA A Suite of Multivariate Analysis tools Sherry Towers SUNY-SB Version 1.0 has been released! useable by anyone with access.
Advertisements

27 th March CERN Higgs searches: CL s W. J. Murray RAL.
Effect of b-tagging Scale Factors on M bb invariant mass distribution Ricardo Gonçalo.
A New Approach to Cluster Finding and Hit Reconstruction in Cathode Pad Chambers and its Development for the Forward Muon Spectrometer of ALICE A.Zinchenko,
Soudan 2 Peter Litchfield University of Minnesota For the Soudan 2 collaboration Argonne-Minnesota-Oxford-RAL-Tufts-Western Washington  Analysis of all.
ROOT courses1 The ROOT System A Data Access & Analysis Framework June 2003 Ren é Brun/EP Histograming & Fitting.
Computing the Posterior Probability The posterior probability distribution contains the complete information concerning the parameters, but need often.
Bo XinRare D Semileptonic Decays 14/23/2006 Studies of Rare Semileptonic D Meson Decays  Introduction  Analysis Technique  Testing the procedure  Results.
1 6 th September 2007 C.P. Ward Sensitivity of ZZ→llνν to Anomalous Couplings Pat Ward University of Cambridge Neutral Triple Gauge Couplings Fit Procedure.
8. Hypotheses 8.4 Two more things K. Desch – Statistical methods of data analysis SS10 Inclusion of systematic errors LHR methods needs a prediction (from.
Statistical Tools PhyStat Workshop 2004 Harrison B. Prosper1 Statistical Tools A Few Comments Harrison B. Prosper Florida State University PHYSTAT Workshop.
1 Nuisance parameters and systematic uncertainties Glen Cowan Royal Holloway, University of London IoP Half.
Search for B     with SemiExclusive reconstruction C.Cartaro, G. De Nardo, F. Fabozzi, L. Lista Università & INFN - Sezione di Napoli.
G. Cowan Lectures on Statistical Data Analysis Lecture 12 page 1 Statistical Data Analysis: Lecture 12 1Probability, Bayes’ theorem 2Random variables and.
1 24 th September 2007 C.P. Ward Sensitivity of ZZ→llνν to Anomalous Couplings Pat Ward University of Cambridge Neutral Triple Gauge Couplings Fit Procedure.
G. Cowan Lectures on Statistical Data Analysis 1 Statistical Data Analysis: Lecture 8 1Probability, Bayes’ theorem, random variables, pdfs 2Functions of.
G. Cowan 2011 CERN Summer Student Lectures on Statistics / Lecture 41 Introduction to Statistics − Day 4 Lecture 1 Probability Random variables, probability.
Chi Square Distribution (c2) and Least Squares Fitting
7/12/2015 Top Pairs Meeting 1 A template fit technique to measure the top quark mass in the l+jets channel Ulrich Heintz, Vivek Parihar.
G. Cowan Lectures on Statistical Data Analysis Lecture 13 page 1 Statistical Data Analysis: Lecture 13 1Probability, Bayes’ theorem 2Random variables and.
G. Cowan Lectures on Statistical Data Analysis Lecture 10 page 1 Statistical Data Analysis: Lecture 10 1Probability, Bayes’ theorem 2Random variables and.
Rao-Cramer-Frechet (RCF) bound of minimum variance (w/o proof) Variance of an estimator of single parameter is limited as: is called “efficient” when the.
G. Cowan Lectures on Statistical Data Analysis 1 Statistical Data Analysis: Lecture 7 1Probability, Bayes’ theorem, random variables, pdfs 2Functions of.
Multivariate Methods of Data Analysis in Cosmic Ray Astrophysics A. Chilingarian, A. Vardanyan Cosmic Ray Division, Yerevan Physics Institute, Armenia.
G. Cowan RHUL Physics Bayesian Higgs combination page 1 Bayesian Higgs combination using shapes ATLAS Statistics Meeting CERN, 19 December, 2007 Glen Cowan.
Atmospheric Neutrino Oscillations in Soudan 2
30th September 2005ROOT2005 Workshop 1 Developments in other math and statistical classes Anna Kreshuk, PH/SFT, CERN.
Distinguishability of Hypotheses S.Bityukov (IHEP,Protvino; INR RAS, Moscow) N.Krasnikov (INR RAS, Moscow ) December 1, 2003 ACAT’2003 KEK, Japan S.Bityukov.
880.P20 Winter 2006 Richard Kass 1 Confidence Intervals and Upper Limits Confidence intervals (CI) are related to confidence limits (CL). To calculate.
A Monte Carlo exploration of methods to determine the UHECR composition with the Pierre Auger Observatory D.D’Urso for the Pierre Auger Collaboration
G. Cowan 2009 CERN Summer Student Lectures on Statistics1 Introduction to Statistics − Day 4 Lecture 1 Probability Random variables, probability densities,
Irakli Chakaberia Final Examination April 28, 2014.
Page 1 Charles Plager LJ+MET, March 23, 2009 Charles Plager UCLA LJ+MET Meeting March 23, 2008 “Throwing PEs” and More.
CHAPTER 2 Statistical Inference, Exploratory Data Analysis and Data Science Process cse4/587-Sprint
Sensitivity to New Physics using Atmospheric Neutrinos and AMANDA-II John Kelley UW-Madison Penn State Collaboration Meeting State College, PA June 2006.
A statistical test for point source searches - Aart Heijboer - AWG - Cern june 2002 A statistical test for point source searches Aart Heijboer contents:
ROOT and statistics tutorial Exercise: Discover the Higgs, part 2 Attilio Andreazza Università di Milano and INFN Caterina Doglioni Université de Genève.
Search for Electron Neutrino Appearance in MINOS Mhair Orchanian California Institute of Technology On behalf of the MINOS Collaboration DPF 2011 Meeting.
1 Methods of Experimental Particle Physics Alexei Safonov Lecture #25.
1 Introduction to Statistics − Day 4 Glen Cowan Lecture 1 Probability Random variables, probability densities, etc. Lecture 2 Brief catalogue of probability.
G. Cowan Lectures on Statistical Data Analysis Lecture 8 page 1 Statistical Data Analysis: Lecture 8 1Probability, Bayes’ theorem 2Random variables and.
Lecture 8 Source detection NASSP Masters 5003S - Computational Astronomy
1 Introduction to Statistics − Day 3 Glen Cowan Lecture 1 Probability Random variables, probability densities, etc. Brief catalogue of probability densities.
A bin-free Extended Maximum Likelihood Fit + Feldman-Cousins error analysis Peter Litchfield  A bin free Extended Maximum Likelihood method of fitting.
G. Cowan Lectures on Statistical Data Analysis Lecture 4 page 1 Lecture 4 1 Probability (90 min.) Definition, Bayes’ theorem, probability densities and.
2005 Unbinned Point Source Analysis Update Jim Braun IceCube Fall 2006 Collaboration Meeting.
G. Cowan Lectures on Statistical Data Analysis Lecture 9 page 1 Statistical Data Analysis: Lecture 9 1Probability, Bayes’ theorem 2Random variables and.
4/12/05 -Xiaojian Zhang, 1 UIUC paper review Introduction to Bc Event selection The blind analysis The final result The systematic error.
A different cc/nc oscillation analysis Peter Litchfield  The Idea:  Translate near detector events to the far detector event-by-event, incorporating.
G. Cowan Lectures on Statistical Data Analysis Lecture 5 page 1 Statistical Data Analysis: Lecture 5 1Probability, Bayes’ theorem 2Random variables and.
Search for exotic contributions to Atmospheric Neutrino Oscillations Search for exotic contributions to Atmospheric Neutrino Oscillations - Introduction.
G. Cowan Lectures on Statistical Data Analysis Lecture 12 page 1 Statistical Data Analysis: Lecture 12 1Probability, Bayes’ theorem 2Random variables and.
G. Cowan Lectures on Statistical Data Analysis Lecture 10 page 1 Statistical Data Analysis: Lecture 10 1Probability, Bayes’ theorem 2Random variables and.
Guglielmo De Nardo for the BABAR collaboration Napoli University and INFN ICHEP 2010, Paris, 23 July 2010.
Kali Calo progress report Dasha Savrina (ITEP/Moscow), Vanya Belyaev.
ROOT: Functions & Fitting Harinder Singh Bawa California State University Fresno.
1 D *+ production Alexandr Kozlinskiy Thomas Bauer Vanya Belyaev
Analysis Tools interface - configuration Wouter Verkerke Wouter Verkerke, NIKHEF 1.
1 Multivariate Histogram Comparison S.I. BITYUKOV, V.V. SMIRNOVA (IHEP, Protvino), N.V. KRASNIKOV (INR RAS, Moscow) IML LHC Machine Learning WG Meeting.
Generalization Performance of Exchange Monte Carlo Method for Normal Mixture Models Kenji Nagata, Sumio Watanabe Tokyo Institute of Technology.
Status of the Higgs to tau tau
Developments in other math and statistical classes
ROOT: Functions & Fitting
Two Interpretations of What it Means to Normalize the Low Energy Monte Carlo Events to the Low Energy Data Atms MC Atms MC Data Data Signal Signal Apply.
Measurement of inclusive branching fraction for y(3686)fX
Rolling Search For a Cascade GRB Signal: 3 Year Results
Computing and Statistical Data Analysis / Stat 7
Top mass measurements at the Tevatron and the standard model fits
Introduction to Statistics − Day 4
Presentation transcript:

Durham1 Collection of algorithms in ROOT Advanced Multivariate & Statistical Techniques FermiLab Workshop May 31, 2002 Ren é Brun CERN

Rene BrunCollection of algorithms in ROOT2 Follow-on of ACAT2000 at FermiLab At ACAT2000, many interesting algorithms were presented. Many of us agreed that it would be nice to collect these algorithms in a coherent framework. Many people have the know-how to develop very clever algorithms, but not the opportunity to make the source/doc available to a large community. I proposed to gradually collect such algorithms in the ROOT suite. Your workshop is an opportunity to make some progress. ACAT2002 in Moscow could be another step.

Rene BrunCollection of algorithms in ROOT3 Which ingredients Statistical methods for Data Analysis Systematics Confidence Limits Signal significance Unfolding techniques Artificial Intelligence Neural nets Pattern recognition techniques Genetic algorithms Wavelet Analysis other algorithms Multi-dim integration, etc

Rene BrunCollection of algorithms in ROOT4 Already in Root Random Number generators standard distributions: flat, gaus, landau, Poisson, from a function 1-d, 2-d, 3-d from histograms 1-d, 2-d, 3-d Principal components Analysis Multidimensional parametrisation and Fitting Histogram peak finders Feldman Cousins Fitting MC fractions to data histogram

Rene BrunCollection of algorithms in ROOT5 FeldmanCousins example void FeldmanCousins() { // Example macro of using the TFeldmanCousins class in root. // Author : Adrian John Bevan // // get a FeldmanCousins calculation object with the default limits // of calculating a 90% CL with the minimum signal value scanned // = 0.0 and the maximum signal value scanned of 50.0 TFeldmanCousins f; // calculate either the upper or lower limit for 10 observerd // events with an estimated background of 3. The calculation of // either upper or lower limit will return that limit and fill // data memebers with both the upper and lower limit for you. Double_t Nobserved = 10.0; Double_t Nbackground = 3.0; Double_t ul = f.CalculateUpperLimit(Nobserved, Nbackground); Double_t ll = f.GetLowerLimit(); cout << "For " << Nobserved << " data observed with and estimated background"<<endl; cout << "of " << Nbackground << " candidates, the Feldman-Cousins method of "<<endl; cout << "calculating confidence limits gives:"<<endl; cout << "\tUpper Limit = " << ul << endl; cout << "\tLower Limit = " << ll << endl; cout << "at the 90% CL"<< endl; }

Rene BrunCollection of algorithms in ROOT6 Generators, Fitting // Quadratic background function Double_t background(Double_t *x, Double_t *par) { return par[0] + par[1]*x[0] + par[2]*x[0]*x[0]; } // Lorenzian Peak function Double_t lorentzianPeak(Double_t *x, Double_t *par) { return (0.5*par[0]*par[1]/TMath::Pi()) / TMath::Max( 1.e-10,(x[0]-par[2])*(x[0]-par[2]) +.25*par[1]*par[1]); } // Sum of background and peak function Double_t myFunction(Double_t *x, Double_t *par) { return background(x,par) + lorentzianPeak(x,&par[3]); } void getrandom() { TCanvas *c1 = new TCanvas("c1","Fitting Demo",10,10,700,900); c1->Divide(1,2); // create a source function with 6 parameters c1->cd(1); TF1 *source = new TF1("source",myFunction,0,3,6); source->SetParameters(-0.86,45.8,-13.3,13.8,0.172,0.987); source->SetNpx(200); source->DrawCopy(); // fill histogram with numbers sampled from the source function c1->cd(2); TH1F *histo = new TH1F("histo","Lorentzian Peak on Quadratic Background",60,0,3); for (Int_t i=0;i<10000;i++) { Double_t sval = source->GetRandom(); histo->Fill(sval); } //fit histogram with original function gStyle->SetOptFit(); histo->Fit(source); }

Rene BrunCollection of algorithms in ROOT7 MonteCarlo Fractions Fitter TFractionFitter: Fits MC fractions to data histogram (see R. Barlow and C. Beeston, Comp. Phys. Comm. 77 (1993) ) The virtue of this fit is that it takes into account both data and Monte Carlo statistical uncertainties. The way in which this is done is through a standard likelihood fit using Poisson statistics; however, the template (MC) predictions are also varied within statistics, leading to additional contributions to the overall likelihood. This leads to many more fit parameters (one per bin per template), but the minimisation with respect to these additional parameters is done analytically rather than introducing them as formal fit parameters.

Rene BrunCollection of algorithms in ROOT8 TFractionFitter example { TH1F *data; //data histogram TH1F *mc0; // first MC histogram TH1F *mc1; // second MC histogram TH1F *mc2; // third MC histogram TObjArray *mc = new TObjArray(3); mc->Add(mc0); // MC histograms are put in this array mc->Add(mc1); mc->Add(mc2); TFractionFitter* fit = new TFractionFitter(data, mc); fit->Constrain(1,0.0,1.0); // fraction 1 to be between 0 and 1 fit->SetRangeX(1,15); // use only the first 15 bins in the fit fit->Fit(); // perform the fit TH1F* prediction = (TH1F*) fit->GetPlot(); data->Draw("Ep"); prediction->Draw("same"); }

Rene BrunCollection of algorithms in ROOT9 Procedure We can give help to create a class in a portable way. Author is responsible for the doc and support We generate the html pages at each release We compile and distribute binaries + source for most machines and compilers. Examples: