A5071802800- Statistical Distribution Estimation and Applications
- Advanced Statistical Methods and Models
- Statistical Methods and Bayesian Inference
- Bayesian Methods and Mixture Models
- Random Matrices and Applications
- Sports Analytics and Performance
- Financial Risk and Volatility Modeling
- Statistical Methods and Inference
- Statistical Mechanics and Entropy
- Sports Performance and Training
- Soil Geostatistics and Mapping
- Geochemistry and Geologic Mapping
- Advanced Combinatorial Mathematics
- Probabilistic and Robust Engineering Design
- Hydrology and Drought Analysis
- Sports injuries and prevention
- Markov Chains and Monte Carlo Methods
- Point processes and geometric inequalities
- Artificial Intelligence in Healthcare and Education
- Complex Systems and Time Series Analysis
- Data Visualization and Analytics
- Cardiovascular Effects of Exercise
- Statistical Methods and Applications
- Probability and Risk Models
- Advanced Mathematical Identities
University of Luxembourg
2022-2025
Cooperative Educational Service Agencies
2023
Ghent University
2015-2022
Ghent University Hospital
2015-2021
Heriot-Watt University
2021
London School of Economics and Political Science
2017
Fonds National de la Recherche
2017
University of Oxford
2017
University of Liège
2017
Université Libre de Bruxelles
2009-2015
Researchers often lack knowledge about how to deal with outliers when analyzing their data. Even more frequently, researchers do not pre-specify they plan manage outliers. In this paper we aim improve research practices by outlining what you need know We start providing a functional definition of then lay down an appropriate nomenclature/classification This nomenclature is used understand kinds can be encountered and serves as guideline make decisions regarding the conservation, deletion, or...
We propose a new general version of Stein's method for univariate distributions. In particular we canonical definition the Stein operator probability distribution which is based on linear difference or differential-type operator. The resulting identity highlights unifying theme behind literature (both continuous and discrete distributions). Viewing as an acting pairs functions, provide extensive toolkit distributional comparisons. Several abstract approximation theorems are provided. Our...
Recent advances in artificial intelligence (AI) present a broad range of possibilities medical research. However, orthopaedic researchers aiming to participate research projects implementing AI-based techniques require sound understanding the technical fundamentals this rapidly developing field. Initial sections primer provide an overview general and more detailed taxonomy AI methods. Researchers are presented with basics most frequently performed machine learning (ML) tasks, such as...
Objective To analyse the association between level of use injury risk estimation feedback (I-REF) provided to athletes and burden during an athletics season. Method We conducted a prospective cohort study over 38-week follow-up period on competing at French Federation Athletics. Athletes completed daily questionnaires their activity, psychological state, sleep, self-reported I-REF use, injuries. for next day, ranging from 0% (no injury) 100% (maximum injury). The primary outcome was...
We provide a new perspective on Stein's so-called density approach by introducing operator and characterizing class which are valid for much wider family of probability distributions the real line. prove an elementary factorization property this propose Stein identity we use to derive information inequalities in terms what call "generalized Fisher distance". explicit bounds constants appearing these several important cases. conclude with comparison between our results known Gaussian case,...
We present 10 different strength-based statistical models that we use to model soccer match outcomes with the aim of producing a new ranking. The are four main types: Thurstone–Mosteller, Bradley–Terry, independent Poisson and bivariate Poisson, their common aspect is parameters estimated via weighted maximum likelihood, weights being importance factor time depreciation giving less weight matches played long ago. Since our goal build ranking reflecting teams’ current strengths, compare on...
Abstract In this work, we propose a new hybrid modeling approach for the scores of international soccer matches which combines random forests with Poisson ranking methods . While forest is based on competing teams’ covariate information, latter method estimates ability parameters historical match data that adequately reflect current strength teams. We compare model to its separate building blocks as well conventional regression models regard their predictive performance all from four FIFA...
Against the COVID-19 pandemic, non-pharmaceutical interventions have been widely applied and vaccinations taken off. The upcoming question is how interplay between social measures will shape infections hospitalizations. Hence, we extend Susceptible-Exposed-Infectious-Removed (SEIR) model including these elements. We calibrate it to data of Luxembourg, Austria Sweden until 15 December 2020. results having highest fraction undetected, Luxembourg infected all three being far from herd immunity...
Introduction Two-thirds of athletes (65%) have at least one injury complaint leading to participation restriction (ICPR) in athletics (track and field) during season. The emerging practice medicine public health supported by electronic processes communication sports represents an opportunity for developing new risk reduction strategies. Modelling predicting the real-time through artificial intelligence using machine learning techniques might represent innovative strategy. Thus, primary aim...
Understanding and quantifying playing styles across different lineups is crucial for evaluating player roles, optimizing lineup combinations, recruiting suitable athletes. This study proposes a comprehensive framework to capture versatility, defined as the range of exhibited various in elite basketball. We collected data from 11,978 games spanning 10 NBA regular seasons extracted lineup-based statistics applied Non-negative Matrix Factorization reduce dimensionality, identifying 6...
Abstract. We develop a method for computing Bayes' factors of conceptual rainfall–runoff models based on thermodynamic integration, gradient-based replica-exchange Markov chain Monte Carlo algorithms and modern differentiable programming languages. apply our approach to the problem choosing from set bucket-type with increasing dynamical complexity calibrated against both synthetically generated real runoff data Magela Creek, Australia. show that using proposed methodology, Bayes factor can...
Skew-symmetric densities recently received much attention in the literature, giving rise to increasingly general families of univariate and multivariate skewed densities. Most those families, however, suffer from inferential drawback a potentially singular Fisher information vicinity symmetry. All existing results indicate that Gaussian (possibly after restriction some linear subspace) play special somewhat intriguing role context. We dispel widespread opinion by providing full...
Data Science is today one of the main buzzwords, be it in business, industrial or academic settings. Machine learning, experimental design, data-driven modelling are all, undoubtedly, rising disciplines if goes by soaring number research papers and patents appearing each year. The prospect becoming a "Data Scientist" appeals to many. A discussion panel organised as part European Conference (European Association for (EuADS)) https://euads.org/edsc/ asked question: "What makes different?" In...
In this paper, we propose tight upper and lower bounds for the Wasserstein distance between any two univariate continuous distributions with probability densities $p_{1}$ $p_{2}$ having nested supports. These explicit are expressed in terms of derivative likelihood ratio $p_{1}/p_{2}$ as well Stein kernel $\tau_{1}$ $p_{1}$. The method proof relies on a new variant Stein's which manipulates operators. We give several applications these bounds. Our main application is Bayesian statistics:...