Seminar

Ruggero Bellio

Ruggero Bellio
(University of Udine)
Bio Ruggero Bellio is Professor of Statistics at the Department Of Economics and Statistics of the University Of Udine. He received the PhD In Statistics from the University of Padova in 2000. His research interests are likelihood methods, mixed models and their applications in various fields, statistical computing, statistical learning. He is very keen on the R statistical software.

Abstract

This talk illustrates a scalable approach to mixed effects modelling with a probit link and a crossed random effects error structure. Random effects with a crossed structure often arise in social and business applications, a notable setting being that of electronic commerce, with random effects related to customers and purchased items, respectively. In sparsely sampled crossed data, the computation for both frequentist and Bayesian estimation can quickly grow superlinearly with respect to the sample size, which severely limits the use of these models for very large settings. The proposed method belongs to the class of composite likelihood estimators and entails the fit of three misspecified reduced models. The resulting estimator is consistent and has an overall computational cost that is linear in the number of observations. The talk will illustrate the method and the results of some numerical experiments, mentioning an alternative approach given by estimation based on stochastic optimisation.

This is a joint work with Art Owen and Swarnadip Ghosh, Stanford University, and Cristiano Varin, Ca’ Foscari University of Venice. Preprint available at arxiv.org/abs/2308.15681

Schedule

  • Friday, March 21, 2025, 11:00-12:00 room seminar 1 @ Povo 0

Details