Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Simulation-Based Inference Evaluation of Tension Between MicroBooNE and MiniBooNE Results in a 3+1 Sterile Neutrino Global Fit

Julia P. Woodward, Joshua Villarreal, John M. Hardin, Austin Schneider, Janet M. Conrad

Abstract

Compatibility between different datasets in a global fit is essential for determining whether a chosen model adequately describes the data. In a 3+1 sterile neutrino global fit, long-standing tensions between datasets sensitive to $ν_e$ appearance and $ν_e/ν_μ$ disappearance indicate a failure of the model to explain the observed data, despite an overall $> 5σ$ improvement over the $3ν$ Standard Model (SM) based on a $χ^2$ fit. Overall, a global preference for the 3+1 sterile-neutrino hypothesis with significant tension between experiments motivates consideration of more complex models, but these are currently computationally prohibitive to evaluate. This paper is the third in a series aimed at reducing computational cost by developing a Simulation-Based Inference (SBI) framework for global fits. Previous papers focused on rapidly fitting the data sets using frequentist (Feldman-Cousins) and Bayesian approaches, while in this work, we formalize a definition of tension within the SBI framework. As an example, we perform a full 3+1 fit to the charged-current quasi-elastic neutrino data from the MiniBooNE experiment and the inclusive neutrino data from the MicroBooNE experiment, located on the same beamline. Using experiment-supplied systematics as is, we find these data sets favor 3+1 at $3.6σ$ and $1.8σ$ respectively, while the tension between the two is $3.3σ$, when fit with the SBI procedure. After correcting for normalization differences between data and Monte Carlo in the MicroBooNE $ν_μ$ samples, the tension relaxes to $2.2σ$, indicating reduced but non-negligible disagreement. The observed tension may reflect both limitations of the 3+1 model in describing the datasets and the presence of systematic effects that impact the experiments differently.

A Simulation-Based Inference Evaluation of Tension Between MicroBooNE and MiniBooNE Results in a 3+1 Sterile Neutrino Global Fit

Abstract

Compatibility between different datasets in a global fit is essential for determining whether a chosen model adequately describes the data. In a 3+1 sterile neutrino global fit, long-standing tensions between datasets sensitive to appearance and disappearance indicate a failure of the model to explain the observed data, despite an overall improvement over the Standard Model (SM) based on a fit. Overall, a global preference for the 3+1 sterile-neutrino hypothesis with significant tension between experiments motivates consideration of more complex models, but these are currently computationally prohibitive to evaluate. This paper is the third in a series aimed at reducing computational cost by developing a Simulation-Based Inference (SBI) framework for global fits. Previous papers focused on rapidly fitting the data sets using frequentist (Feldman-Cousins) and Bayesian approaches, while in this work, we formalize a definition of tension within the SBI framework. As an example, we perform a full 3+1 fit to the charged-current quasi-elastic neutrino data from the MiniBooNE experiment and the inclusive neutrino data from the MicroBooNE experiment, located on the same beamline. Using experiment-supplied systematics as is, we find these data sets favor 3+1 at and respectively, while the tension between the two is , when fit with the SBI procedure. After correcting for normalization differences between data and Monte Carlo in the MicroBooNE samples, the tension relaxes to , indicating reduced but non-negligible disagreement. The observed tension may reflect both limitations of the 3+1 model in describing the datasets and the presence of systematic effects that impact the experiments differently.
Paper Structure (23 sections, 6 equations, 8 figures, 1 table)

This paper contains 23 sections, 6 equations, 8 figures, 1 table.

Table of Contents

  1. Introduction
  2. Inputs to the Three-Oscillation-Channels
  3. The 3+1 Model
  4. The MiniBooNE Experiment
  5. The MicroBooNE Experiment
  6. Generating Pseudo-data
  7. Simulation-Based Inference Framework
  8. Frequentist Fitting Strategy with Simulation-Based Inference
  9. Computing PG Tension with Simulation-Based Inference
  10. ML Penalization for Unmodeled Systematic Effects
  11. Fits Using Simulation-Based Inference
  12. Fit to the MiniBooNE Data Set
  13. Fit to the MicroBooNE Data Sets
  14. Joint Fits to MiniBooNE and MicroBooNE Data Sets
  15. Fit Significances
  16. ...and 8 more sections

Figures (8)

  • Figure 1: (Top) MiniBooNE $\nu_e + \bar{\nu}_e$ data and data/MC ratio distributions, with the Standard Model prediction shown in blue and the SBI-computed best fit 3+1 prediction in pink. Statistical error bars are taken from MiniBooNE:2022emn. (Bottom) SBI-computed MiniBooNE 3+1 fit results with the Wilks'-based $95\%$ CL from MiniBooNE:2022emn overlaid for comparison.
  • Figure 2: Same as Fig. \ref{['fig:mB']} but for the MicroBooNE $\nu_\mu$ FC dataset, without any additional nuisance parameters added to the fit.
  • Figure 3: SBI-computed fit results for the MicroBooNE dataset with additional nuisance parameters added to the fit in the $\nu_\mu$ samples. We overlay the Wilks'-based $95\%$ CL from MiniBooNE:2022emn for comparison.
  • Figure 4: SBI-computed joint fit results for MiniBooNE and MicroBooNE, with the $95\%$ CL from MiniBooNE:2022emn overlaid for comparison.
  • Figure 5: Normalized histogram of $\tilde{\chi}^2_{PG}$. The blue histogram is constructed from $50,000$ null realizations, and the observed value is shown as the dashed line. Relevant $\chi^2$ distributions are overlaid in red.
  • ...and 3 more figures