Luiza Piancastelli (University College Dublin)

will speak on

Multivariate Conway-Maxwell-Poisson Distribution: Sarmanov Method and Doubly-Intractable Bayesian Inference

Time: 12:00PM
Date: Mon 29th November 2021
Location: Online [map]

Abstract: A multivariate count distribution with Conway-Maxwell (COM)-Poisson marginals is proposed having desirable features such as (i) it admits a flexible covariance matrix allowing for both negative and positive non-diagonal entries; (ii) it overcomes the limitation of the existing bivariate COM-Poisson distributions in the literature that do not have COM-Poisson marginals; (iii) it allows for the analysis of multivariate counts. Inferential challenges for this model arise from its intractable likelihood that is based on the Sarmanov (1966) method. We handle Bayesian inference for the resulting doubly-intractable posterior with an approximate exchange algorithm specification and a Grouped Independence Metropolis-Hastings. The proposed model is applied to model a real data application on the numbers of goals scored by the home and away teams in the Premier League (2018 to 2021), illustrating the potential of the MultCOMP model and providing insight on the effect of a lack of crowds during the COVID-19 pandemic on the home team advantage.

Reference: Piancastelli, L.S.C., Friel, N., Barreto-Souza, W. and Ombao, H. (2021). Multivariate Conway-Maxwell-Poisson Distribution: Sarmanov Method and Doubly-Intractable Bayesian Inference. ArXiv link: https://arxiv.org/abs/2107.07561

Join the Zoom call: https://ucd-ie.zoom.us/j/68316324831

(This talk is part of the Working Group on Statistical Learning series.)

PDF notice

Return to all seminars