Robust test statistics for data sets with missing correlation information

Abstract

Not all experiments publish their results with a description of the correlations between the data points. This makes it difficult to do hypothesis tests or model fits with that data, since just assuming no correlation can lead to an over- or underestimation of the resulting uncertainties. This work presents robust test statistics that can be used with data sets with missing correlation information. They are exact in the case of no correlation and either guaranteed to be conservative -- i.e. the uncertainty is never underestimated -- in the presence of correlations, or they are also exact in the degenerate case of perfect correlation between the data points.

arXiv comment24 pages, 15 figures, 2 tables, 2 code listings

References in corpus (8)

• The GENIE Neutrino Monte Carlo Generator
• First Measurement of the Muon Neutrino Charged Current Quasielastic Double Differential Cross Section
• First Measurement of the Muon Anti-Neutrino Double-Differential Charged Current Quasi-Elastic Cross Section
• The NEUT Neutrino Interaction Simulation
• Extensions of Superscaling from Relativistic Mean Field Theory: the SuSAv2 Model
• NUISANCE: a neutrino cross-section generator tuning and comparison framework
• Measurement of Quasielastic-Like Neutrino Scattering at $⟨E_{\nu }⟩\sim 3.5$~ GeV on a Hydrocarbon Target
• Measurement of the muon anti-neutrino double-differential cross section for quasi-elastic scattering on hydrocarbon at~$E_{\nu }\sim 3.5$ GeV

Cited by in corpus (3)

• New CC0π GENIE Model Tune for MicroBooNE
• Parametrized uncertainties in the spectral function model of neutrino charged-current quasielastic interactions for oscillation analyses
• Hypothesis tests and model parameter estimation on data sets with missing correlation information