More details coming soon

27/11/2026 - 11:00 - ISBA C115 
 

Ivan Kojadinovic  

(Université de Pau et des Pays de l'Adour) 

Will give a presentation on : 

An algorithmic procedure for solving the generalized minimum information checkerboard copula problem

Abstract: 

The minimum information copula principle (see Meeuwissen and Bedford, 1997) is a maximum entropy-like approach for finding the least informative copula that satisfies a certain number of expectation constraints specified either from expert knowledge or the available limited data. In this presentation, we first propose a generalization of this principle allowing the inclusion of additional constraints fixing certain higher-order margins of the copula. We next prove that the associated optimization problem has a unique solution under a natural condition. As the latter problem is intractable in general, following the existing literature, we consider its version with all the probability measures involved in its formulation replaced by checkerboard approximations. This amounts to attempting to solve a
so-called discrete I-projection linear problem. We then use the seminal results of Csiszar (1975) to derive an IPFP-like procedure for solving the latter and provide theoretical guarantees for its convergence. We next report numerical experiments in dimensions up to four with substantially finer discretizations than those encountered in the literature. We finally discuss the computational cost of the procedure and hint at memory saving implementations that should allow to tackle higher dimensional problems under certain conditions.

More details coming soon

11/12/2026 - 14:30 - ISBA C115 
 

Alessandra Luati  

(Imperial College London) 

More details coming soon

13/11/2026 - 14:30 - ISBA C.115
 

Kellie Archer 

(The Ohio State University) 

Will give a presentation on : 

High-Dimensional Mixture Cure Models for Time-to-Event Data 

Abstract: 

Medical breakthroughs in recent decades have led to cures for many diseases including cancer.  For example, various groups have shown that advances in therapy for leukemia and myelodysplastic syndrome have increased overall survival rates and some groups have identified factors related to long-term survival, defined as survival exceeding three years, which included that patients treated on newer treatment regimens were more likely to be long-term survivors. In fact, some argue that these improved outcomes indicate that some AML patients can be considered “potentially cured.” The mixture cure model (MCM) is a time-to-event model that is used when a cured fraction exists. MCMs assume the population consists of two subgroups, those cured will not experience the event of interest and those susceptible to the event of interest. Therefore, “cured” can be considered synonymous with attaining long-term relapse-free survival. Thus, there are two regression components in MCMs which permit identification of features associated with cure and/or latency of susceptible patients. Many researchers have sought to identify a prognostic model for a time-to-event outcome. When the covariate space is high-dimensional, as in the case with gene expression data, typically penalized Cox proportional hazards (PH) models are fit. However, when a dataset includes a cured fraction, MCMs are more appropriate than Cox PH models. In these scenarios, Cox models often yield inaccurate hazard and survival estimates due to violations of the proportional hazards assumption. Despite their utility, robust methods for fitting MCMs to high-dimensional data remain scarce. In this talk, I will present our recent work extending MCMs to effectively handle high-dimensional covariate spaces, offering a more accurate prognostic tool for modern high-dimensional datasets.

26/02/2027 - 14:30 - ISBA C115 
 

Debarghya Ghoshdastidar

(Technical University of Munich) 

More details coming soon

05/03/2027 - 14:30 - ISBA C115 
 

Emilie Devijver

(Université Grenoble Alpes) 

will give a presentation on 

Causal inference, causal abstraction and causal machine learning

Abstract:

Causal inference has been introduced to move beyond simple correlations, offering capabilities for explainability, generalization and action. 
In this talk, I will present two main axis developped within our team: 
1/ causal abstraction, as a practical means to encode expert knowledge in the form of concepts (rather than variables) and to apply this knowledge effectively. To do so, I will provide a formalism that enables causal inference while accounting for cycles;
2/ out-of-distribution generalization, and how to leverage the causal structure to make predictions in a new environment through the causal invariance concept.

This work is in collaboration with Marianne Clausel, Eric Gaussier, Théotime Le Goff, Adèle Ribeiro and Clément Yvernes.

03/12/2026 - 14:30 - ISBA CV9 / C045
 

Thai Nguyen

(Université Laval) 

Will give a presentation on : 

Pareto and Bowley Reinsurance Games in Peer-to-Peer Insurance

Abstract: 

We propose a peer-to-peer (P2P) insurance scheme comprising a risk-sharing pool and a reinsurer. A plan manager determines how risks are allocated among members and ceded to the reinsurer, while the reinsurer sets the reinsurance loading. Our work focuses on the strategic interaction between the plan manager and the reinsurer, and this focus leads to two game-theoretic contract designs: a Pareto design and a Bowley design, for which we derive closed-form optimal contracts. In the Pareto design, cooperation between the reinsurer and the plan manager leads to multiple Pareto-optimal contracts, which are further refined by introducing the notion of coalitional stability. 

In contrast, the Bowley design yields a unique optimal contract through a leader-follower framework, and we provide a rigorous verification of the individual rationality constraints via pointwise comparisons of payoff vectors. Comparing the two designs, we prove that the Bowley-optimal contract is never Pareto optimal and typically yields lower total welfare. In our numerical examples, the presence of reinsurance improves welfare, especially with Pareto designs and a less risk-averse reinsurer. We further analyze the impact of the single-loading restriction, which disproportionately favors members with riskier losses.

05/02/2027 - 09:00 - ISBA C115 - 
 

09:00  Introduction

09:10  Kamal Gasser
"Clustering extremal dependence structures via Wasserstein distances between angular measures"

Abstract:
Grouping random vectors by their joint extremal behaviour is a recurring task in climate science, hydrology, and finance. Classical clustering is driven by typical observations, while existing extremal procedures either summarize dependence by a single coefficient or discretize the simplex in an ad hoc way. We propose a dissimilarity that compares complete angular measures by optimal transport: each angular measure is estimated above a high radial threshold, mapped from the simplex into a Euclidean space of Aitchison coordinates, and compared by the 1-Wasserstein distance. The methodology is fully nonparametric and requires no fitted model. The dissimilarities feed any distance-based clustering algorithm. Under intermediate-threshold asymptotics and mild moment and separation conditions, hierarchical clustering and partitioning around medoids based on them recovers the latent partition with probability tending to one. In simulations, the proposed approach matches established competitors when clusters differ mainly in the strength of extremal dependence and performs substantially better when equal-strength clusters differ in the shape of their angular measures, especially when asymmetry drives the separation. Applications to rain--snowmelt--runoff extremes in Swiss catchments and to extremal co-movement between international equity markets and the US market reveal hydrologically and economically interpretable groups.

09:50

10:20  

10:50  Break

11:10

11:50

12:20

12h30  Lunch 

04/12/2026 - 14:30 - ISBA C.115
 

Hongyuan Cao 

(University of Wisconsin-Madison) 

Will give a presentation on : 

Kernel meets sieve for transformed hazards model with sparse longitudinal covariates

Abstract: 

We study the transformed hazards model with time-dependent covariates observed intermittently for the censored outcome. Existing work assumes the availability of the whole trajectory of the time-dependent covariates, which is unrealistic. We propose combining kernel-weighted log-likelihood and sieve maximum log-likelihood estimation to conduct statistical inference. The method is robust and easy to implement. We establish the asymptotic properties of the proposed estimator and contribute to a rigorous theoretical framework for general kernel-weighted sieve M-estimators. Numerical studies corroborate our theoretical results and show that the proposed method performs favorably over competing methods. The analysis of a data set from a COVID-19 study in Wuhan identifies clinical predictors that otherwise cannot be obtained using existing methods.