CREST will host an afternoon workshop on November 6th, 2024 aimed at equipping PhD students (first year and more) with essential digital skills for research. This session will cover practical topics including an introduction to the command line interface, open-source practices with Git and GitHub (Émilien Schultz, and LaTeX for scientific publishing (Nicolas Chopin). No prior experience is required.
Market Power, Randomization and Regulation, Simon Loertscher (University of Melbourne)
|
SCHEDULE |
Monday |
9th September 2024
16th September 2024 |
From 9:00 to 12:15
From 13:30 to 16:45 |
Room 2003 |
|
Thursday |
12th September 2024 |
From 13:30 to 16:45 |
Room 2003 |
Outline
Monopoly and monopsony pricing problems are of long-standing interest in economics. With the
emergence of large online platforms in the digital age, they have received renewed salience and attention.
This course provides an overview of recent theoretical advances based on incomplete information models.
These shed new light on a broad array of issues, ranging from the optimality of rationing and involuntary
unemployment to resale, opaque pricing by multi-product monopolies, market definition in antitrust and
merger effects to optimal regulation in the tradition of Ramsey.
The incomplete information approach not only provides a coherent explanation for otherwise puzzling
phenomena. For example, why does the seller of concert tickets underprice these, thereby inviting resale,
rather than set a market-clearing price? Why do multi-product sellers offer opaque products, leaving buyers
in the dark as to what they purchased until after the fact? It also implies that, because of incentive
compatibility, perfect price discrimination is not possible, and it permits clear-cut predictions of the form
price discrimination takes, with and without resale. By assuming that the firm always chooses the optimal
mechanism, the approach is immune to the criticism that its predictions hinge on ad-hoc restrictions. A
tradeoff between social surplus and profit derives from the primitives.
Estimation of Functionals of High-Dimensional Parameters: Bias Reduction and Concentration, Vladimir Koltchinskii (Georgia Institute of Technology)
|
SCHEDULE |
Monday |
6th May 2024 13th May 2024 |
From 13:00 to 16:15 |
Room 2033 |
|
Thursday |
16th May 2024 23rd May 2024 |
From 13:00 to 16:15 |
Room 2033 |
Aims and objectives
The aim of this course will be on a circle of problems related to estimation of real valued functionals of parameters of high-dimensional statistical models. In such problems, it is of interest to estimate onedimensional features of a high-dimensional parameter that are often represented by nonlinear functionals of certain degree of smoothness defined on the parameter space. The functionals of interest could be estimated with faster convergence rates than the whole parameter (sometimes, even with parametric rates). The examples include, for instance, such problems as estimation of linear functionals of principal components (that are nonlinear functionals of unknown covariance) in high-dimensional PCA. The goal is to discuss several mathematical methods that provide a way to develop estimators of functionals of highdimensional parameters with optimal error rates in classes of functionals of some Hölder smoothness.
Moreover, when the degree of smoothness of the functional is above certain threshold, the estimators in question have parametric √𝑛 error rate and are asymptotically efficient, whereas the error rates become slower than √𝑛 when the degree of smoothness is below the threshold.
The following topics will be covered (at least, to some extent):
• preliminaries in high-dimensional probability and analysis (concentration inequalities, comparison inequalities, Hölder smoothness of operator functions, etc);
• non-asymptotic bounds and concentration inequalities for sample covariance in high-dimensional and dimension-free frameworks;
• some approaches to concentration inequalities for smooth functionals of statistical estimators;
• higher order bias reduction methods in functional estimation;
– methods based on Taylor expansion and estimation of polynomials with reduced bias;
– iterative bias reduction and bootstrap chains;
– linear aggregation of plug-in estimators with different sample sizes and jackknife estimators;
• minimax lower bounds in functional estimation (applications of van Trees inequality, Nemirovski’s construction of least favorable functionals, etc);
• Examples:
– high-dimensional and infinite dimensional Gaussian models: functionals of mean and of covariance;
– log-concave models, in particular, log-concave location families;
– high-dimensional exponential families;
– nonparametric models, functionals of unknown density;
– linear functionals of spectral projections of matrix parameters.
Paula GOBI (Université Libre de Bruxelles) – “t.b.a.”
Applied Micro Seminar : Every Tuesday
Time: 12:15 pm – 13:30 pm
Date: 19th of March
Room : 3001
Paula GOBI (Université Libre de Bruxelles) – “t.b.a.”
Abstract :
Organizers:
Benoît SCHMUTZ (Pôle d’économie du CREST)
Clément MALGOUYRES (Pôle d’économie du CREST)
Sponsors:
CREST
An Introduction to Conformal Prediction and Distribution-Free Inference, Rina Foygel Barber (University of Chicago)
|
SCHEDULE |
Monday |
18th March 2024 |
From 13:00 to 16:15 |
Room 2033 |
|
Thursday |
21st March 2024 |
From 13:00 to 16:15 |
Room 2033 |
Aims and objectives
This short course will introduce the framework of distribution-free statistical inference, and will provide an in-depth overview of both theoretical foundations and practical methodologies in this field. We will cover methods including holdout set methods, conformal prediction, cross-validation based methods, calibration procedures, and more, with emphasis on how these methods can be adapted to a range of settings to achieve robust uncertainty quantification without compromising on accuracy. The course will also introduce the theoretical results behind these methods, including the role of exchangeability (and its variants) in establishing the distribution-free validity of these methods, as well as more classical theoretical results establishing how these distribution-free methods relate to the answers we would obtain via parametric models or other classical assumption-based techniques. Our theoretical overview will also cover hardness results that carve out the space of inference questions that are possible or impossible to answer within the distribution-free framework.
Christos MAKRIDIS (Arizona State University) “Learning from Friends in a Pandemic: Social Networks and the Macroeconomic Response of Consumption”
Macro seminar
Time : 12h15 – 13h30
Date : 18 Mars 2024
Salle 3001
Christos MAKRIDIS (Arizona State University) “Learning from Friends in a Pandemic: Social Networks and the Macroeconomic Response of Consumption”
Abstract: Households often learn about the macroeconomy through social communications. We show how communication via social networks allows idiosyncratic shocks to propagate into meaningful macroeconomic responses. We first motivate such a mechanism by showing the responses of daily consumption spending of U.S. counties to plausibly exogenous variations in their social-network exposure to the COVID-19 conditions of geographically remote regions. Various identification strategies confirm that the detected consumption responses were primarily through the channel of expectations, rather than preference interdependence or physical infection risks. Then, we incorporate a belief formation mechanism through social networks into an otherwise standard heterogeneous-agent consumption model. We show that a pandemic-augmented version of this model, where infections initially hit a fraction of more connected regions and gradually propagated via social networks, produces macroeconomic dynamics more aligned with the empirical patterns of aggregate consumption and cross-sectional variations. We also show that such a mechanism is more relevant when the initial shocks hit the more connected agents and when the influences are more dispersed in the network.
Organizer : Julien PRAT (CREST)
Dynamic Factor Models, Matteo Barigozzi (Università di Bologna)
|
SCHEDULE |
Monday |
4th March 2024 11th March 2024 |
From 13:00 to 16:45 |
Room 2033 |
|
Thursday |
7th March 2024 |
From 13:00 to 16:15 |
Room 2033 |
Aims and objectives
The aim of this course is to provide an introduction to factor models in time series analysis by teaching students the basic theoretical foundations and by illustrating them some applications to macroeconomics and finance.
In the last years large datasets have become increasingly available to researchers and practitioners in many disciplines. In particular, during this big data revolution the analysis of high-dimensional time series has become one of the most active subjects of modern statistical methodology with applications in the most different areas of science including finance, econometrics, meteorology, genomics, chemometrics, complex physics simulations, biological and environmental research. Although the value of information is unquestionable, the possibility of extracting meaningful and useful information out of this large amount of data is also of great importance. To achieve such dimension reduction, several new analytical and computational techniques have been developed under the name of machine learning methods. Among these factor models not only are one of the pioneering methods in the field of unsupervised learning (dating back to Spearman, 1904), but up to these days have also been one of the most popular and most employed ones.
We start by discussing principal component analysis as a useful dimension reduction technique for large panels of time series. This is the most simple example of factor model (the static model) which we then generalize to include all temporal relations among the considered variables (the dynamic model). We then focus on the case in which the dynamic model can be re–written as a state space model and we present its estimation via Kalman filter and the Expectation Maximization algorithm. We then consider application of these models in two fields. First, in macroeconometrics for building indicators of the business cycle, for nowcasting, and for policy analysis problems. Related to these we briefly discuss how to deal with the issues of non–fundamentalness and cointegration. Second, in financial econometrics for volatility modelling and forecasting. Related to these we briefly discuss the complementarity of factor and network models and the issue of conditional heteroskedasticity. Real–data applications taken from existing papers are discussed during the lectures. Matlab or R code will be provided.