Fully Bayesian Unfolding
classification
⚛️ physics.data-an
keywords
bayesianunfoldingfullymethodagostiniappliedbeforechoosing
read the original abstract
Bayesian inference is applied directly to the problem of unfolding. The outcome is a posterior probability density for the spectrum before smearing, defined in the multi-dimensional space of all possible spectra. Regularization consists in choosing a non-constant prior. Despite some similarity, the fully bayesian unfolding (FBU) method, presented here, should not be confused with D'Agostini's iterative method.
This paper has not been read by Pith yet.
Forward citations
Cited by 1 Pith paper
-
Empirical-Bayes Unfolding of $\gamma$-ray Spectra
Empirical-Bayes hierarchical unfolding for gamma-ray spectra with Poisson ON/OFF likelihood, adaptive Richardson-Lucy prior, and NUTS posterior sampling, yielding spectra consistent with frequentist regularized ML.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.