Program – Provisional: Multivariate Dependence Modeling Including Copula and Covariance Stucture Modeling

David Wood

  •  22 Feb 2027 - 26 Feb 2027
     9:00 am - 5:00 pm
Organisers:

Pavel Krupskiy (University of Melbourne)
Clara Grazian (University of Sydney()

Program Description:

This workshop focuses on two powerful and complementary approaches to modelling multivariate dependence: copula models and covariance structure modelling. Together, these methodologies provide a robust foundation for understanding and interpreting the relationships within complex, high-dimensional data. Copula models have emerged as a powerful and flexible framework for capturing complex dependence structures in multivariate data. Their key advantage lies in decoupling the modelling of marginal distributions from the joint dependence structure, allowing for highly adaptable and realistic multivariate modelling. This versatility has led to widespread adoption across diverse domains, including finance, insurance, hydrology, climatology, ecology, and geology.

In this workshop, we will explore cutting-edge developments in copula-based modelling, focusing on methods designed to handle large-scale datasets. Emphasis will be placed on recent advances in inference tools and their application to real-world data, showcasing how these models can reveal relationships that traditional approaches often miss. As a complementary perspective, the workshop also highlights covariance structure modelling, a powerful approach for capturing dependencies, particularly in spatio-temporal data such as climate, environmental, and geological observations. Accurate estimation of the mean structure is crucial in this context, and we will delve into bi-linear mean models (tensor models) while jointly modelling the covariance structure. Together, these two methodological frameworks — copulas and covariance models — provide a comprehensive toolkit for tackling the challenges of modern multivariate data analysis.

Program Structure:

TBD

Registration:

  • Deadline: TBD
  • Registration is by invitation only. If you are interested to participate in this program, please contact one of the organisers with your CV and research background.
  • Arrival date: 21 February 2026

ASSOCIATED EVENTS

MATRIX Wine and Cheese Afternoon 23 February 2027
On the first Tuesday of each program, MATRIX provides a pre-dinner wine and cheese afternoon. Produce is locally-sourced to showcase delicacies from the region.