Estimation of Voigt Distribution Parameters: A Bayesian Approach

Cannas, Massimo

;Piras, Nicola

2025-01-01

---

Abstract

The Voigt profile, the density obtained from the convolution of a Gaussian and a Cauchy, is widely used in atomic and molecular spectroscopy. We exploit a characterization of the Voigt profile as a location mixture of the Cauchy distribution for parameter estimation via Gibbs sampler. A simulation study comparing the performance of the proposed algorithm against two widely used algorithms is presented. The proposed approach exhibits a promising performance in terms of MSE.