A regularization-independent Information Criterion for correlated data
Davide Laudicina
Abstract
We generalize the Akaike Information Criterion (AIC) to allow model and data subset selection when regularized fits are performed on correlated data. While the standard generalization to regularized fits is inherently dependent on the type of regularization applied to the data covariance matrix, our definition, the RIAIC, is, on average, regularization-independent. The regularization-independence is achieved by defining the new Information Criterion as an estimator of the model-dependent contribution to the correlated Kullback--Leibler divergence. The Bayesian extension is derived as well. We test the new criterion on lattice QCD data by performing regularized fits and show that it reproduces the selection properties of the standard correlated AIC, with fluctuations which are compatible with the expected statistical uncertainties. These results enable the application of subset and model selection techniques to cases where regularizations are applied to the covariance matrix of the fitted dataset.
Create a lesson
Related papers
Review of Pion and Nucleon Gluon Parton Distribution Functions from Lattice QCD
William Good, Huey-Wen Lin
Half-strings across the deconfinement transition of a U(1) quantum link model
Indrajit Sau, Arnab Sen, Debasish Banerjee
Equation of state of hot and dense QCD at the physical point via the complex Langevin equation
Michael Mandl, Dénes Sexty, Daniel Unterhuber
Testing the D234 Dirac operator
Jorge Baeza-Ballesteros, Andreas Risch, Mattis Roost et al.
Light fermions can trigger a Brout-Englert-Higgs effect
Axel Maas, Sofie Martins, Georg Wieland
Interpretable uncertainties for bounds on spectral functionals from Euclidean correlators
Christoph Lehner