Dalitz plot analysis of the \(D^{+}\rightarrow K^{-}K^{+}K^{+}\) decay

Sebastián Ordoñez-Soto

Universidad Nacional de Colombia

 

Supervisors: Alberto C. dos Reis and Diego Milanés

 

November 4th, 2022

 

IJCLab LHCb group meeting

Contents

  • Introduction
  • Data sample
  • Data analysys 
  • Preliminary results
  • Outlook

Introduction

Motivation

  • Decays of \(D\) mesons into three pseudoscalars are excellent laboratories to study light-meson spectroscopy.
  • In a \(D\rightarrow K^{-}K^{+}K^{+}\) decay, the \(K\bar{K}\) scattering amplitudes can be determined.
    • Need phenomenological models.

Problem

  • No tools for a description from first principles.
  • The challenge:
    • Dynamics: Heavy-quark weak transitions, hadron formation and final-state interactions (FSI)
    • Strong part: effects of coupled channels, three-body interactions and isospin.

Data sample

Original sample

2016-Down

  • Data from the Run 2 of the LHCb is used in this analysis. The corresponding Monte Carlo samples are also available.

(GeV)

(GeV)

(GeV)

(MeV)

Invariant mass distribution  of the \(D\) candidates

Dalitz plot  of the \(D\) candidates

Original sample

2016-Down

(GeV)

(GeV)

\(s_{K^{-}K^{+}}\)  projection of the Dalitz

\(s_{K^{-}K^{+}}\)  projection of the Dalitz

Data sample

Original sample

2016-Up

(GeV)

(MeV)

Dalitz plot  of the \(D\) candidates

Data sample

Invariant mass distribution  of the \(D\) candidates

Original sample

2016-Up

(GeV)

(GeV)

Data sample

Original sample

2017-Down

(GeV)

(MeV)

Dalitz plot  of the \(D\) candidates

Data sample

Invariant mass distribution  of the \(D\) candidates

Original sample

2017-Down

(GeV)

(GeV)

Data sample

Original sample

2017-Up

(GeV)

(MeV)

Dalitz plot  of the \(D\) candidates

Data sample

Original sample

2017-Up

(GeV)

(GeV)

Data sample

Original sample

2018-Down

(GeV)

(MeV)

Dalitz plot  of the \(D\) candidates

Data sample

Original sample

2018-Down

(GeV)

(GeV)

Data sample

Original sample

2018-Up

(GeV)

(MeV)

Dalitz plot  of the \(D\) candidates

Data sample

Original sample

2018-Up

(GeV)

(GeV)

Data sample

Data analysis: Pre-selection

Pre-selected sample

2016-Down

  • Preselection cuts include PID (pi_ProbNNk) and clone tracks  (slope variables)

(GeV)

(GeV)

(MeV)

Dalitz plot  of the \(D\) pre-selected candidates

Mass distribution  of the \(D\)  pre-selected candidates

Preselected sample

2016-Down

(GeV)

(MeV)

Mass distribution  of the \(D\)  pre-selected candidates

\(s_{K^{-}K^{+}}\)  projection of the pre-selected sample

Data analysis: Pre-selection

Data analysis: Pre-selection

Preselected sample

2016-Up

Dalitz plot  of the \(D\) pre-selected candidates

Mass distribution  of the \(D\)  pre-selected candidates

Data analysis: Pre-selection

Preselected sample

2016-Up

Mass distribution  of the \(D\)  pre-selected candidates

\(s_{K^{-}K^{+}}\)  projection of the pre-selected sample

Data analysis: Pre-selection

Preselected sample

2017-Down

Data analysis: Pre-selection

Preselected sample

2017-Down

Data analysis: Pre-selection

Preselected sample

2017-Up

Data analysis: Pre-selection

Preselected sample

2017-Up

Data analysis: Pre-selection

Preselected sample

2018-Down

Data analysis: Pre-selection

Preselected sample

2018-Down

Data analysis: Pre-selection

Preselected sample

2018-Up

Data analysis: Pre-selection

Preselected sample

2018-Up

Data analysis: MC-matching

Fits of the preselected sample

2016-Down (Fold 1)

2016-Down (Fold 2)

MC matching: Splot weights

Data analysis: MC-matching

2016-Down (Fold 1)

Signal weights

Background weights

MC matching: Splot weights

Data analysis: MC-matching

2016-Down: Fold 1

Invariant mass distribution weighted by sWeights

Gradient boosted weights

2016-Down (Fold 1)

  • Monte Carlo and data distributions before the reweighting

Data analysis: MC-matching

Gradient boosted weights

2016-Down (Fold 1)

Data analysis: MC-matching

Kolmogorov-Smirnov before reweighting

Kolmogorov-Smirnov after reweighting

Gradient boosted weights

  • Monte Carlo and data distributions after the reweighting

2016-Down (Fold 1)

Data analysis: MC-matching

Input data for MVA

2016-Down (Fold 1)

Data analysis: MVA

Reweighted Monte Carlo is used as signal

Data sidebands are used as background

(MeV)

(MeV)

MVA training results

Discriminant variables

Data analysis: MVA

MVA training results

Correlation matrices

Data analysis: MVA

MVA training results

Overtraining check

Figures of merit

Data analysis: MVA

MVA application results

Data (Fold 2)

Monte Carlo (Fold 2)

Data analysis: MVA

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > -1.0

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > -0.9

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > -0.7

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > -0.5

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > -0.3

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > -0.1

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > 0.1

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > 0.3

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > 0.5

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > 0.7

Data analysis: Figures of merit

(MeV)

(MeV)

Optimal cut value

Data (Fold 2)

Monte Carlo (Fold 2)

BDTG_val > 0.9

Data analysis: Figures of merit

(MeV)

(MeV)

Fits after BDTG cut

BDTG_val > 0.9

BDTG_val > 0.8

Data analysis: Figures of merit

BDTG_val > 0.7

BDTG_val > 0.6

Data analysis: Figures of merit

Fits after BDTG cut

Monte Carlo efficiency

Data analysis: Figures of merit

Signal efficiency and purity

Data analysis: Figures of merit

Significance

Data analysis: Figures of merit

Purity x Significance

Data analysis: Figures of merit

Significance as figure of merit

Optimal cut: BDTG_val = 0.0

Data analysis: Final distributions

(MeV)

Mass distribution  of the \(D\)  selected candidates

Optimal cut: BDTG_val = 0.0

(MeV)

(MeV)

Data analysis: Final distributions

Mass distribution  of the \(D\)  selected candidates (Signal region)

Mass distribution  of the \(D\)  selected candidates (Background region)

Significance as figure of merit

Significance as figure of merit

Optimal cut: BDTG_val = 0.0

(GeV)

(GeV)

(GeV)

Data analysis: Final distributions

Dalitz plot  of the \(D\) selected candidates in the signal region

Optimal cut: BDTG_val = 0.0

(GeV)

(GeV)

(GeV)

Data analysis: Final distributions

Dalitz plot  of the \(D\) selected candidates in the sidebands

Significance as figure of merit

Thank you!