Skip to main navigationSkip to main content
The University of Southampton
Southampton Statistical Sciences Research Institute

S3RI Seminar - "Bayesian inference for bivariate copulas with additive models for dependence, marginal location, scale and shape: an application in paediatric ophthalmology", Prof Mario Cortina Borja (University College London) Seminar

S3RI Seminar
14:00 - 15:00
6 December 2018
Lecture Theatre 7B, Building 54, Mathematical Sciences, University of Southampton, Highfield Campus, SO17 1BJ

For more information regarding this seminar, please email Dr Helen Ogden at .

Event details

Motivated by data on visual acuity from a large sample of children aged between 3 and 8 years, we propose bivariate copula models with dependence parameters, and sinh-arcsinh marginal densities with location, scale and shape parameters that depend on a covariate through additive models. We perform inference about the unknown quantities of our model in the Bayesian framework using a Markov chain Monte Carlo algorithm. We apply our model to paediatric ophthalmic data to gain new insights about the processes which cause changes in visual acuity with respect to age, including the age-related nature of the copula dependence parameter. We analyse predictive distributions to identify children with unusual sight characteristics, distinguishing those who are bivariate, but not univariate outliers. In this way we provide an innovative tool that enables clinicians to identify children with unusual sight who may otherwise be missed.

Work in collaboration with Julian Stander and Luciana Dalla Valle (Plymouth),  Brunero Liseo (Rome), Charlotte Taglioni (Padua), and Angie Wade (UCL).

Speaker information

Professor Mario Cortina Borja, University College London. Mario is an applied statistician with a background in Actuarial Science and Mathematics; he has worked in many scientific areas. His present research involves: Epidemiology of infectious disease, particularly HIV and Hepatitis C; Modelling physical activity using accelerometry data; Seasonal patterns, especially birthdays and the effect of date of birth on adult health; Record linkage methods; Modelling discrete, long-tailed data; Modelling multivariate circular responses; Transformations in bounded outcome variables and Applications of copulae models.

Privacy Settings