Likelihood Asymptotics in Nonregular Settings: A Review with Emphasis on the Likelihood Ratio

by Prof. Alessandra Rosalba Brazzale (Università di Padova)

Wednesday 26 Mar 2025, 11:00 → 12:00 Europe/Zurich

40/S2-A01 - Salle Anderson (CERN)

This seminar will offer a  review the most common situations in which the regularity conditions that underlie classical likelihood-based parametric inference fail, focusing on the large-sample properties of the likelihood ratio statistic.  Three main classes of problems emerge: boundary problems, indeterminate parameter problems —- which include non-identifiable parameters and singular information matrices -- and change-point problems.  I'll emphasise analytical solutions, consider software implementations where available, and summarise how the key results are derived.

Alessandra R. Brazzale is Associate Professor of Statistics at the Department of Statistical Sciences of the University of Padova.  She received her PhD from the Swiss Federal Institute of Technology, Lausanne in 2000.  Since then, she has specialized in developing and implementing statistical methods and models for likelihood-based and likelihood-type inference and higher-order asymptotic theory, mainly with application to the Life Sciences.  Brazzale received the John M. Chambers statistical software award in 2001.  She furthermore co-authored a 2007 CUP monograph on the application of higher-order asymptotic theory.  During the past decade, she developed an interest in statistical applications to Particle and Astrophysics, with emphasis on nonregular settings.

Coffee will be served at 10:30.

CERN_nonreg_talk.pdf