听厂辫辞苍蝉辞谤蝉听听 |
|---|
听
EOS Partnership KULeuven - UCLouvain 2021-2025
Actuaries and STatisticians endeavour to design innovative, inclusive insurance products in a changing RISK landscape (ASTeRISK)
Actuaries design and value risk transfers by analyzing insurance data sets with a sophisticated statistical toolbox. Current practice faces technical challenges in the design and estimation of risk models from fine-grained data. At the same time, trust in modern-day insurance is under pressure. Policyholders and regulators expect value creation from the collected granular data in the form of better insurance products and a wider coverage of (new types of) risks. New forms of mutual insurance (e.g., peer-to-peer covers) recently emerged outside traditional insurance, inspired by evolutions in the sharing economy. In response to these challenges our project aims to shape the statistical toolbox necessary for building valid, reproducible and viable risk models for the granular data collected in a variety of key actuarial tasks. Building on solid methodological foundations our project innovates by explicitly incorporating a dynamic prediction and prevention view into these risk models. Moreover, we aim to contribute to a better, more inclusive and fair access to insurance with the design of a wide array of new risk-sharing solutions and the development of novel quantitative tools to spot (proxy) discrimination in insurance pricing. Finally, the project targets tailored statistical models to assess actual riskiness and to refine the implementation of the recent 鈥榬ight to be forgotten鈥� that should improve the access of cancer survivors to essential insurance products.
Promoter: Katrien Antonio (KU Leuven) and Donatien Hainaut (UCLouvain)
Website :
UCLouvain FSR 2021-2023
Graph informed sufficient dimension reduction: a bridging conditional independence framework for distributed estimation and inference
The proposed project focuses upon conditional independence modelling and provides new methodological aspects for extending the estimation of probabilistic graphical models (PGMs) while performing supervised feature extraction methods proposed in the statistics literature. This will be performed by exploiting the close connection between these two distinct modelling frameworks and by retaining the strong points of each framework and thus making feasible the identification of a central dimension reduction space in a high-dimensional setting. As such, the estimation of PGMs in settings where datasets are stored on distributed clusters will be of central interest.
Promotor: E. Pircalabelu, Researcher: Ensiyeh Nezakati
F.R.S.-FNRS (Fonds de la Recherche Scientifique) 2021-2025
Learning structures and patterns from multivariate extremes
Learning new knowledge from data at the outer regions of a sample is challenging because, by definition, the amount of such data is scarce and model assumptions have a disproportionately large impact on the conclusions. Our overall objective is to leverage and develop techniques from multivariate extreme value analysis to enrich the tool-set for analyzing and learning from the extremes of a set of data. Specifically, we aim to extend current models for multivariate peaks-over-thresholds to the case that such peaks occur in small clusters of variables and we aim to develop methods to learn the graphical structure of dependence relations between the extremes of the components of a high-dimensional random vector. The methods developed will be put into practice in the analysis of data involving weather and climate extremes.
Promoters : Anna Kiriliouk (UNamur) and Johan Segers
ETHIAS Chair 2019-2023
Fully funded Pension Systems
The purpose of this interdisciplinary research project (law, actuarial science) is to look at the future of fully funded occupational pension schemes in the context of ageing and low interest rates.
Promoter: Pierre Devolder
BNB (Banque nationale de Belgique) 2020-2023
Control variates for simultaneous Monte Carlo integration of a large number of functions
Integrals of functions of several variables arise in many fields of applied mathematics: statistics, machine learning, mathematical finance, and so on. Monte Carlo methods aim to calculate such integrals by sampling integration points randomly. One of the main variance-reduction techniques in Monte Carlo integration uses control variates. The method is particularly well suited to handle a large number of integrands in parallel. The aim of the project is to develop probabilistic theory for control-variate methods for Monte Carlo integration and to use this theory to design better methods. The results to be shown take the form of concentration inequalities, convergence rates, and asymptotic distributions of the random integration error. The focus is on the case where there is a large number of integrands and a large number of potential control variates.
Promoter : Johan Segers
UCLouvain FSR 2021-2023
Processus de L茅vy fractionnaires et applications en finance math茅matique (2021-2023)
Promotor: D. Hainaut, Researcher: Jean-Loup Dupret
Partnership KULeuven - UCLouvain 2020-2024
Quantile regression for censored data
One of the statistical challenges in survival analysis is the study of the relationship between a time-to-event response T and a set covariates X. This can be done using a wide variety of regression techniques like, for example, linear, AFT or Cox models. A robust and flexible alternative to these classical models is quantile regression, which has gained considerable popularity and interest in recent years. Many methods have been developed for quantile regression with completely observed data. But when data are subject to censoring, statistical estimation and inference become more difficult, and the literature is sparse. The existing work focuses on the case of i.i.d. data with a right-censored response, but in practice censoring mechanisms can be quite complicated (e.g. interval censoring) and may concern both the response and the covariates. The objective of this project is to develop and study consistent and computationally efficient procedures to conduct estimation and inference in quantile regression models with complicated censoring mechanisms. To this end, an enriched asymmetric Laplace distribution will be proposed and studied. Once studied, this distribution will be used to investigate the case of quantile regression with (1) censored response, (2) censored covariates and (3) censored response and censored covariates.
Promoter: Anouar El Ghouch & Ingrid Van Keilegom
FW-B (Federation Wallonie-Bruxelles) 2020-2025
Imperfect Data : From Mathematical Foundations to Applications in Life Sciences (IMAL)
We are witnessing a period of time where the data collection potential has increased exponentially. The cautionary tale of this big data era is that large amounts of data do not necessarily contribute to an increment in our knowledge about the underlying phenomenon. One of the principal reasons for this is that even though one would desire to measure a characteristic for a subject, in many instances one can only get an approximate measurement due to difficulty in obtaining the direct measurement of the desired phenomenon (e.g. tumor size), non-replicability across instances (e.g. blood pressure), necessity to obtain numerous measurements rapidly, sometimes at the cost of accuracy. As a result, many modern observed markers are proxies for the real data because invasive, costly or too complex methods would be required to obtain accurate measurements. In this project we study how one can correct for different types of imperfect data when building statistical models with a focus on applications coming from life sciences. Imperfect data appear in different contexts, structures and models, and this project focuses on two common settings which regularly suffer from imperfect data: data in a regression context with imperfectly measured explanatory variables (Theme 1) and highdimensional or functional data with measurement error (Theme 2).
Promoters : Catherine Legrand (porte-parole, UCLouvain), Anouar El Ghouch (UCLouvain), Philippe Lambert (UCLouvain / ULi猫ge),
Eugen Pircalabelu (UCLouvain), Germain Van Bever (UNamur), Ingrid Van Keilegom (UCLouvain / KU Leuven).
ARC project
FW-B (Federation Wallonie-Bruxelles) 2018-2023
Sustainable, Adequate and Safe Pensions
This interdisciplinary research project (law, economics, actuarial science, philosophy) aims at critically assessing the key conditions that a public pension system should fulfil to be successfully reformed. Our hypothesis is that there are three such conditions: i) financial sustainability, ii) social adequacy and iii) safe governance. Hence, the 鈥楽AS鈥� acronym. Our goal is to identify the pension architecture that is the most likely to generate SAS pensions.
Promoters : Pierre Devolder, Alexia Autenne, Jean Hindriks, Vincent Vandenberghe, Axel Gosseries (ARC project)
FW-B (Federation Wallonie-Bruxelles) 2018-2023
Negative and ultra-low interest rates: behavioral and quantitative modelling
Interest rates are a cornerstone of economics and finance. They are at the foundation of asset pricing and monetary policy, and more generally of all intertemporal choices made by market participants and institutions every day, with huge consequences for the economic activity and wellbeing of our societies. Until recently, it was assumed (mostly implicitly) that interest rates could only possibly be positive. Notwithstanding, in the wake of the financial crisis initiated in 2008, major central banks of developed countries have been brought to conduct rates policies that turned them negative. The consequences of such a paradigm shift are both potentially huge and not well understood yet. This research project aims at shedding light on these consequences, both from an academic and a policy viewpoint, following three intertwined research lines that bring together a multidisciplinary team of researchers working on Behavioral Finance, Macro Finance, and Quantitative Finance.
Promoters: Catherine D鈥橦ondt, Julio D谩vila, Leonardo Iania, Christian Hafner, Olivier Corneille and Frederic Vrins.
ARC project
SAS Partnership 2018-2022
SAS
The SAS software is one of the most used statistical software in the world. Since several years, there exist a partenariat between SAS and Institut de Statistique, Biostatistique et Sciences Actuarielles (ISBA) through which courses of programming in SAS and data mining techniques are organized. These courses are open to all master students as well as to PhD students and to all researchers of the UCLouvain. Within the context of this partenariat, SAS also support (financially and logistically) the organisation of short courses within ISBA.