Information détaillée concernant le cours

[ Retour ]

École d’hiver 2022


6-9 février 2022

Responsable de l'activité

Sebastian Engelke


Prof. Sebastian Engelke UNIGE

Mme Caroline Gillardin Coordinatrice CUSO


Prof. Jose Blanchet, Stanford University (USA) (online)

Prof. Julie Josse, Institut national de recherche en sciences et technologies du numérique (France) (online)

Prof. Leonhard Held, University of Zurich (en présentiel ave Mme Micheloud qui l'assistera, PhD)



Trois cours de 4.5 heures chacuns seront donnés par des spécialistes internationaux sur des sujets de pointes en statistiques.

Prof. Jose Blanchet de Stanford University (USA)

title: "Distributional robustness/Optimal Transport Statistics"

Abstract : Optimal transport has received substantial attention in statistics and machine learning, especially in recent years, in order to design robust estimators. This short course will briefly introduce basic notions in optimal transport theory (e.g. the Wasserstein distance, the Moonge-Kantorovich problem and fundamental duality notions). Then, we will explain how these results are used for distributionally robust performance analysis. For example, for risk calculations involving stochastic processes, and other worst case expectations. We then introduce min-max games in which the statistician wishes to minimize a certain loss against and adversary (nature) which explores the impact of overfitting using optimal transport. Thus, automatically endowing estimators with good generalization properties. We study statistical properties of these estimators, including asymptotic normality and optimal (in a natural sense) confidence regions. We finish our discussions with computational considerations.


Updated LECTURE 08.02.2022

The lectures will largely be based on the following tutorial paper, which contains many references: (this has been published by INFORMS tutorials here:

Prof. Julie Josse de l' Institut national de recherche en sciences et technologies du numérique (France)

title: "Overview of methods for handling missing values"

Abstract: Missing values are ubiquitous in the practice of data analysis.In this series of lectures, we will start by presenting classical methods for handling missing data (simple imputation, multiple imputation, likelihood-based methods) developed in an inferential framework, where the objective is to best estimate parameters and their variance in the presence of missing data.We will emphasize very powerful methods of simple and multiple imputation based on low-rank approximations that can be applied to heterogeneous data (quantitative, categorical).

We will then present recent results in a supervised learning framework. A striking result is that naive imputation strategies (such as mean imputation) can be optimal, as the supervised learning method does the hard work. The fact that such a simple approach can be relevant may have important consequences in practice.

We will also discuss how missing value modeling can be easily incorporated into tree models, such as gradient boosted trees, resulting in a learner that has been shown to perform very well, including in challenging non-random missingness settings.Notebooks will be presented. Finally, we will quickly present how such results are useful in the context of causal inference with missing values in the covariates


Prof. Leonhard Held, Univesity of Zurich. Program interactif.

title : "Design and Analysis of Replication Experiments - with an introduction to the R package ReplicationSuccess"

Abstact :Replication studies are increasingly conducted in order to confirm original findings. However, there is currently no consensus on how to design such studies and to define replication success. The purpose of this tutorial is to describe and compare statistical approaches for the design and analysis of replication studies. The standard method based on significance is discussed, as well as alternative approaches based on effect sizes or meta-analysis. Participants will learn how to use the R-package ReplicationSuccess and will apply the methods to real data from large-scale replication projects. Prerequisites include basic knowledge of R and familiarity with standard concepts of statistical inference.Trainers:
Leonhard Held is Professor of Biostatistics at the University of Zurich and Director of the Center for Reproducible Science. Charlotte Micheloud is a PhD student developing statistical methodology for the design and analysis of replication studies.






Program (will be adjusted according to the speakers)



Sunday 06.02

Monday 07.02

Tuesday 08.02

Wednesday 09.02



Leonhard Held + Charlotte Micheloud

Leonhard Held + Charlotte Micheloud

Julie Josse



Coffee Break

Coffee Break

Coffee Break



Leonhard Held + Charlotte Micheloud

Julie Josse

Julie Josse



 Lunch or pique-nique

 Lunch or pique-nique

 Lunch or pique-nique




 Comité scientifique (15h-17h)



Welcome tea

Coffee Break

Coffee Break



Jose Blanchet

Jose Blanchet

Jose Blanchet







Dinner (raclette)




Les Diablerets, Eurotel (2G) + online


Eurotel Victoria

1865 Les Diablerets (VD)

Accès aux Diablerets :

EN VOITURE Autoroute A9, direction Grand St-Bernard, sortie Aigle. Puis la route Aigle - Les Diablerets - Col du Pillon (20km). 

EN AVION Aéroports internationaux de: - Genève (120 km) - Zürich (250 km) - Bâle (200 km)

EN TRAIN  (HORAIRE DES TRAINS - RAILWAY TIMETABLE)  International TGV Paris - Lausanne. En hiver, TGV des Neiges Paris - Lausanne - Aigle.

Swiss Train schedule : From: Geneva airport, To: Les Diablerets, gare. Trains directs jusqu'à Aigle. Ensuite train de montagne A.S.D (Aigle - Sépey - Diablerets) Durée des trajets: Lausanne - Aigle (30 minutes), Aigle - Les Diablerets (50 minutes).

Visa pour la Suisse (Swiss Online Visa application)

Météo en suisse (

Adresse salle de gym Maison des congrès : Chemin des Grandes Isles, 1865 Ormont-Dessus (Entre la gare les Diableret et l'Eurotel)


Tarifs si l'activité a lieu en présentiel :

Doctorant CUSO chambre double: 200 CHF

Doctorant CUSO chambre simple: 350 CHF

Post-doctorant CUSO chambre double: 300 CHF

Post-doctorant CUSO chambre simple: 450 CHF

Professeur CUSO chambre double: 400 CHF

Professeur CUSO chambre simple: 550 CHF

Non CUSO universitaire chambre double: 850 CHF

Non CUSO universitaire chambre simple: 1000 CHF

Non CUSO privé chambre double: 1300 CHF

Non CUSO privé chambre simple: 1500 CHF

Lors de votre inscription, merci de bien vouloir indiquer dans la zone commentaire si vous désirez une chambre simple, ou double et le nom de la personne avec qui vous souhaiteriez partager votre chambre. Dans le cas où rien n'est indiqué, une chambre simple sera réservée.


IMPORTANT : vous devrez présenter à votre arrivée à l'hôtel un certificat covid valide 2G (personnes vaccinées ou guéries). Nous vous founirons des autotest gratuits en vous demandant de vous tester à votre arrivée à l'hôtel, puis chaque matin. Nous vous demandons également d'amener suffisament de masques pour toute la durée de votre séjour.

Versement sur compte postal (seulement si l'activité à lieu en présentiel):

CCP 12-1873-8
IBAN : CH0509000000120018738.
Merci d'écrire votre nom suivi du no" 22220002 " lors du paiement. Thank you to write your name on the payment wording and nb" 22220002".

IMPORTANT : si vous vous inscrivez, et que vous avez un empêchement, veuillez nous contacter dès que possible. Sinon le prix de toutes le nuitées à l'hôtel vous seront intégralement facturées si l'hôtel le demande. If you register and are unable to attend, please contact us as soon as possible. Otherwise you might be billed for all nights in the hotel.

 IMPORTANT: si vous vous inscrivez pour participer par visio conference, merc de bien vouloir avertir le secrétariat ([email protected]) que vous ne viendrez pas en personne



Délai d'inscription 29.01.2022
short-url short URL

short-url URL onepage