Advanced Statistical Seminar — NMST611
(Summer Term 2026)
When: Wednesdays, 15:40 - 17:00
Where: Lecture Room Praktikum KPMS
The Advanced Statistical Seminar consists of presentations delivered (typically in person) by invited foreign speakers or guests of the Department of Probability and Mathematical Statistics. Assorted topics from modern statistics — theory and applications — are usually communicated during the talks.
Seminar schedule
- 18.02.2026 | 15:40 | Diego Bolón Rodríguez
Université Libre de Bruxelles, Belgium
Title: A review on highest density region estimation
Highest density regions (HDRs for short) are sets where the density function of the data exceeds a given (and usually high) threshold. Estimating the HDRs of a population from a data sample is a useful tool for data visualisation, cluster analysis, and outlier detection. Due to its practical utility, HDR estimation for Euclidean data has been widely considered in the literature. However, HDR estimation in other contexts has only recently been addressed. In this talk, we begin by exploring the different techniques that have been developed for HDR estimation in the Euclidean context. This introduction allows us to highlight the particular issues of this specific topic, such as how to measure consistency in this context. Then, we explore the recent efforts to extend these techniques for populations supported on a manifold, including a novel approach that combines a density function estimator with some a priori geometric information. This seminar is based in a joint work with Rosa M. Crujeiras and Alberto Rodríguez-Casal.
- 04.03.2026 | 15:40 | Irène Gijbels
KU Leuven, Belgium
Title: Semiparametric regression for circular response
In standard settings, with real-valued random variables, there are many tools to estimate conditional mean, mode, quantiles, variance functions, ranging from parametric, semiparametric and nonparametric procedures (depending on realistic assumptions that can be justified or not). In many applications the variable of interest, the response, is circular, for example flight orientation of migrating birds, ocean current direction, orientation of geological fractures, orientation of proteins in structural biology. Possible predicting variables can be real-valued (flight altitude of birds, windspeed, ...) or also of circular type (e.g. wind direction). Moreover, predicting factors can have an influence not only on centrality measures (mean direction, mode, quantiles, ...) but also on the variability (variance, dispersion, concentration, ...). Motivated by a study on flight orientations of migrating birds, using parametric models, and facing limitations of these, we propose a method to estimate semiparametrically the conditional modal direction as well as conditional concentration. We introduce the semiparametric regression model, and present results for statistical inference. The added value of the developed semiparametric method is illustrated in an ecology application.
This talk is based on joint work with J. Ameijeiras-Alonso (University of Santiago de Compostela). - 18.03.2026 | 15:40 | Michele Cavazzutti
Charles University, Czech Republic
Title: Nonparametric methods for the analysis of spatial and functional data over non-convex multidimensional supports
In recent decades, the widespread use of advanced measurement technologies has enabled the collection of spatial and functional data defined over non-convex multidimensional domains. Such data often exhibit irregular sampling, large missing regions, and reflect spatially dependent phenomena that may be only partially known a priori. In this talk, we explore the problem of performing uncertainty quantification within the framework of spatial regression with partial differential equation (PDE) regularization, which addresses these challenges by incorporating prior physical knowledge through a roughness penalty. This class of models is flexible and encapsulates information on the complex geometry of the domain via a finite element discretization of the problem. However, the aforementioned roughness penalty also induces bias in the model, making uncertainty quantification challenging in this setting. We therefore present several parametric and nonparametric inference approaches to address these issues, discussing the resulting statistical properties in terms of control of the Type I error and the power of the resulting hypothesis tests. Furthermore, we show how the same geometrically compliant methodology developed for the semiparametric regression setting can be extended to the context of integrated functional data depth, enabling the ranking of functional data that are partially observed over multidimensional supports, including Riemannian manifolds. This talk is based on two works conducted in collaboration with Prof. Laura Sangalli (Politecnico di Milano) and Prof. Eleonora Arnone (Università di Torino).
- 01.04.2026 | 15:40 | Bruno Ebner
Karlsruhe Institute of Technology, Germany
Title: TBA
...
- 15.04.2026 | 15:40 | Valentina Massaroto
Leiden University, The Netherlands
Title: TBA
...
- 29.04.2026 | 15:40 | Katarzyna Szczerba
University of Luxembourg, Luxembourg
Title: TBA
...
- 13.05.2026 | 15:40 | Stanislav Volgushev
University of Toronto, ON, Canada
Title: TBA
...
Advanced Statistical Seminar (Archiv)
|
The archive (seminar speakers, talk titles, and the abstracts) of the Advanced statistical seminar (NMST611)
from previous semesters together with the title of the talks and short abstracts. |
|
| Winter term 2025/2026 | Summer term 2025 |
|---|---|
| Winter term 2024/2025 | Summer term 2024 |
| Winter term 2023/2024 | Summer term 2023 |
| Summer term 2022 | |
| Summer term 2021 | |
