The Luxembourg Statistical Society is organizing its second LSS price ceremony.

This event will award an LSS price for the best internship report or master thesis of the University of Luxembourg in the domain of statistics.

The aim of this event is to reinforce the importance of statistics as an applied discipline and bring people together with a shared interest in applied statistics. It awards the students that did an exceptional work contributing to statistics via their internship report.

When and where? This event will take place on the 29th of January 2026 at 6 pm at Campus Belval in room MSA 3530.

Save the date!

Who are the nominees?

• Clara Duchossois, Master in Data Science: Machine Learning and Data Analytics in Human Resources: Absenteeism Prediction and Strategic HR Management (Internship done at Banque de Luxembourg)
• Ifeoluwapo Favour Edun, Master in Mathematics: Automatic Classification of IT Help Desk Tickets via LDA, K-means and BERTopic Modeling (Academic Master Thesis)
• Alexandros Meichanetzidis, Master in Financial Mathematics: Probabilistic Shadow Valuation of Maritime Assets in an Illiquid-Alternative Investment Fund (Internship done at JP Morgan)

Program:

• 18:00: Welcome words by Jang Schiltz, president of the LSS and introduction to the LSS prize by Katarzyna Szczerba, scientific jury president.
• 18:10: Presentations by the Nominees
• 19:00: Jury Deliberation
• 19:15: Announcement of the Winner

If you are interested, please register to this event by sending an email to luxstatsoc@gmail.com.

We are happy to announce that the talk by Professor Masanobu Taniguchi from Waseda University, Japan will take place on Wednesday April 17 at 6.30pm at the University of Luxembourg on the Kirchberg Campus (6, Rue Richard Coudenhove-Kalergi) in the main building, room D10F. The talk will be followed by a drink.

Abstract

This talk delves into the transformative and innovative potential of two cutting-edge technologies – Large Language Models (LLMs) and blockchain – for the field of statistics, and in particular, Official Statistics.

In this talk, topics from the elementary statistics to modern new one will be delivered. First, we explain the meaning of mean, variance, deviation, of statistical data. As the theoretical aspect, we introduce mathematical statistics, mentioning estimator, unbiasedness, sufficiency etc. For goodness of estimator, Rao-Blackwell theorem is stated with an interesting episode of C.R.Rao.

Then we explain the elements of time series, feature of financial data, and nonlinear time series models. Their optimal inference is discussed based on LAN structure, and as an optimal estimator, MLE is given. We apply time series analysis for medical data e.g., ECoG, EMG data to illuminate diseases.

Actual data are essentially non-stationary. For such time series, the optimal inference is established. Classification problems are addressed for such data.

We also deal with the problem of portfolio. For this the optimal portfolio coefficients are given by our modern statistical methods.

As an interesting episode, Nightingale’s life will be mentioned. We explain her relationship to statistics and graphical method in statistics.

Finally, Granger causality will be introduced two time series, and we apply this concept for tourism statistics.

The Luxembourg Statistical Society would like to invite you to the talk by Cristiano Tessitore on “Large Language Models and Blockchain in Statistics: Innovations and Use Cases”. The talk will take place on Thursday 7^th March at 12:00 PM.

Raul Ian Sosa is from Eurostat

Abstract

This talk delves into the transformative and innovative potential of two cutting-edge technologies – Large Language Models (LLMs) and blockchain – for the field of statistics, and in particular, Official Statistics.
LLMs, with their enormous data processing and natural language understanding capabilities, have gained increasing popularity since the launch of OpenAI’s ChatGPT in November 2022. This development has made the generative AI sector increasingly competitive, with new breakthroughs emerging daily, making LLMs a perfect “teammate” for scientists.
Blockchain technology, thanks to its innovative features like immutability, transparency, and decentralization, can strengthen trust in official statistical results.
By combining LLMs and blockchain, this speech aims at showcasing a future where statistical analysis is more efficient, accurate, and trustworthy. Attendees will gain insights into practical applications and the potential of these technologies to transform the statistical landscape

Practical information

Date: 7 March 2024
Time: 12:00-13:00
Place: Webinaire
Fee: free
Registration: please register by sending an email to Christophe Ley (Société Statistique Luxembourgeoise)

The Luxembourg Statistical Society would like to invite you to the talk by Raul Ian Sosa on “Understanding the Potentials and Pitfalls of ChatGPT and AI Language Models”. The talk will take place on Wednesday 13^th September at 6:00 PM.

Raul Ian Sosa is doctoral researcher at Department of Physics and Materials Science of University du Luxembourg

Abstract

In recent months, AI language models have skyrocketed in popularity among both companies and end users. ChatGPT, in particular, has become the fastest-growing web platform of all time and serves as a formidable tool for tasks ranging from simple text summarization to complex ones such as writing entire emails and generating pieces of software. In this talk, we will delve into the creation and evolution of this revolutionary technology, demystifying what these AI models are and envisioning the future they might shape. Despite these new solutions offering immense possibilities, they also pose challenges such as the risk of generating misleading information and perpetuating societal biases. We will dissect these issues, understand their roots, and explore strategies that we, as users, can employ to mitigate them, ensuring the responsible use of these sophisticated tools.

Practical information

Date: 13 September 2023
Time: 18:00-19:00
Place: Metz Room, ground floor of the SnT Building (29, av, JF Kennedy, Luxembourg)
Fee: free
Registration: please register by sending an email to Christophe Ley (Société Statistique Luxembourgeoise)

The Luxembourg Statistical Society would like to invite you to the talk by Monica Pratesi on “Towards a framework to harness data by citizens: the Istat experience”. The talk will take place on Monday 17^th April at 12:00 PM in virtual mode.

Monica Pratesi is Director of the Department for Statistical Production, Istat, Italian National Institute of Statistics

Abstract

Big data, smart statistics, digital administrative data, and citizens are inseparable: from smartphones, meters, fridges and cars to internet platforms, the data of most digital technologies is citizen data, that is the data of the citizens and on citizens.

Their importance and their involvement ask for a thorough rethinking of the relations between the citizens and the production of official statistics. In the session we will focus on examples of the reuse of data generated by citizens on themselves (in processes like using services, making telephone calls or when they populate administrative archives during their lives as students, workers, retired persons etc.) to produce Official Statistics. Examples will use the statistical production process from the Permanent Population and Households Census, from the integration process of utility data files (e.g. electricity, gas) in the design of different activities, like for instance the Household Energy Consumption Survey or data collection on violence against women and on discrimination (LGBTQ+ population, Roma and homeless) conducted in cooperation with civil society organizations, also touching upon the issue of the reuse of privately-held-data in official statistics.

Practical information

Date: 17 April 2023
Time: 12:00-13:00
Place: online
Fee: free
Registration: please register by sending an email to Christophe Ley (Société Statistique Luxembourgeoise)

The Luxembourg Statistical Society would like to invite you to the talk by Prof. Christophe Ley on “Statistical Modelling and Data Science”. The talk will take place on Thursday 20^th April at 17:30 PM in person mode. It will be in the room 3.330 of the Maison du Savoir on the University Campus Belval.

Christophe Ley joined the Department of Mathematics at the University of Luxembourg in November 2021 as Associate Professor. He is an applied mathematician working in the field of statistics/data science, with a strong taste for interdisciplinary work. Before joining the University of Luxembourg, he was Associate Professor at Ghent University in Belgium. He obtained his PhD from the Université libre de Bruxelles in 2010, where he had also studied except for the first year, which he did at the University of Luxembourg during its first year of existence in 2003-2004.

He is President of the Luxembourg Statistical Society, President of the European organisation for Advanced Courses in Statistics (ECAS), and Leader of the international network S-TRAINING (new data science methods for sports analytics and sports medicine). Among his achievements are the Prix Marie-Jeanne Laurent-Duhamel, rewarding the best PhD thesis in Statistics over a period of 3 years among all French-speaking universities world-wide, and an Elected Membership at the International Statistical Institute. 

He leads the research team MIDAS which stands for Modelling – Interdisciplinary – Data (Science) – Applied – Statistics. The team’s aim is to develop innovative statistical and machine learning procedures based on new mathematical and computational tools to tackle the challenges posed by modern days’ increasingly complex and large data sets.

Abstract

Since the turn of the millenium, technological innovations have allowed the collection and storage of increasingly complex and large data sets, a situation that has in turn spurred significant advances in such domains as medicine, meteorology, renewable energies, engineering, sports, social sciences, and many more. Most of these scientific advances are grounded on strong mathematical foundations – see for instance the variety of epidemiological models required to understand the COVID-19 spread, or the statistical theories underpinning clinical studies. In his talk, Prof. Christophe Ley will provide a friendly overview on statistical modelling and data science, two important themes of his own research and that of his research team. He will describe modern challenges and, in particular, focus on research he wishes to undertake in the coming years.

Practical information

Date: 20 April 2023
Time: 17:30-20:00
Place: room 3.330, Maison du Savoir (Belval campus)
Fee: free
Registration: please register by sending an email to Christophe Ley (Société Statistique Luxembourgeoise)

The Luxembourg Statistical Society would like to invite you to the talk by Prof. Sarah Gelper on “Estimating the Effect of Second Screening on TV Viewing using Mobile Diaries”. The talk will take place on Thursday 23^rd February at 12:30 PM in hybrid mode. If you attend it in person, it will be in the room on the first floor of the Maison du Nombre, MNO-E01-0146030, on the University Campus Belval.

Sarah Gelper is Associate Professor in Business Analytics with a Specialization in Marketing Analytics at Université du Luxembourg.

Abstract

People often use second devices, such as their mobile or tablet, while watching TV. Such « second screening » might enhance engagement and increase viewership but also risks distracting viewers and reducing interest in the show. Estimating the effect of second screening on repeat viewing is challenging for three reasons. (1) We need to observe behavior on two different devices (the data challenge); (2) the effect of second screening might vary with viewer and show characteristics (the heterogeneity challenge); and (3) second screening is highly endogenous behavior (the endogeneity challenge). This paper leverages large-scale individual-level data from mobile diaries of 1,702 US TV viewers on 2,755 primetime TV shows over 3-6 weeks in which viewers report both viewing and second screening. The data are augmented to account for a broad set of marketing influences such as show communications and promos. We characterize heterogeneity by four latent viewer classes (heavy viewers, sports enthusiasts, sitcom viewers, and light viewers) and use a causal forest analysis to estimate the heterogeneous effects of second screening on repeat viewing. Despite the possible risk of distraction, second screening does not harm actual TV viewing, and this effect is robust over various types of second screening activities and across viewer segments. Second screening impacts attitudes more than behaviors – unlike actual viewing, stated viewing intentions are affected right after the second screening experience. Specifically, viewing intentions increase when people use a second screen to communicate about the show, but decrease when they use the second screen for show-unrelated activities.

Practical information

Date: 23 February 2023
Time: 13:00-14:00
Place: room MNO-E01-0146030 (Belval campus)
Fee: free
Registration: please register by sending an email to Christophe Ley (Société Statistique Luxembourgeoise)

The Luxembourg Statistical Society would like to invite you to the talk by Prof. Jean-Michel Poggi on “Random Forests: Introduction and industrial applications”. The talk will take place on Thursday 27^th October at 1:00 PM and will be hosted in room MNO 1.040 (Belval campus).

Jean-Michel Poggi is Professor of Statistics at Université Paris Cité and member of the Lab. Maths Orsay (LMO) at Université Paris-Saclay, in France.
His research interests are in nonparametric time series, wavelets, tree-based methods (CART, Random Forests, Boosting) and applied statistics. Research activities combine theoretical and practical contributions together with industrial applications (mainly environment and energy) and software development.
He is Associate Editor of three journals: Journal of Statistical Software (JSS), Advances in Data Analysis and Classification (ADAC) and Journal of Data Science, Statistics, and Visualisation (JDSSV).
He is President of the European Network for Business and Industrial Statistics (ENBIS). He was President of the French Statistical Society (SFdS) from 2011 to 2013, and President of the European Courses in Advanced Statistics (ECAS) from 2015 to 2017.

Abstract

Random forests (RF) are a statistical learning method extensively used in many fields of application, thanks to its excellent predictive performance. RF are part of the family of tree-based methods and inherit intrinsic flexibility of trees: adapted to both classification and regression problems, easily extended to many different types of data.
We will first focus on an introduction of RF, the definition of the variable importance measures and the related variable selection capability. We then illustrate their practical power in two different applied contexts. The first one relates to physiological signal processing and addresses the functional variable selection for driver’s stress level classification. The second is about the aggregation of multi-scale experts for bottom-up electricity load forecasting.

Practical information

Date: 27 October 2022
Time: 13:00-14:00
Place: room MNO 1.040 (Belval campus)
Fee: free
Registration: please register by sending an email to Christophe Ley (Société Statistique Luxembourgeoise)