Teaching

I teach a complete statistical learning track for master students in economics. The course is designed for a total of 75 hours, including practical sessions at the end of each chapter. Chapters are divided into three categories : theory, descriptive methods, predictive methods.  Numbers indicate what I believe to be an optimal ordering. Students are expected to know basic econometrics (OLS, causal inference)

THEORY :

• T1 : A crash course in statistical theory (1)

• T2 : Reminders in analysis and algebra (2)

• T3 : Introduction to the ERM framework (7)

• T4 : Mathematical foundations of supervised learning (9)

• T5 : Numerical optimization (10)

DESCRIPTIVE METHODS :

• D1 : Dimensionality reduction (3)

• D2 : Clustering (5)

• D3 : Statistical analysis of network data (6)

PREDICTIVE METHODS :

• P1 : Logistic regression (4)

• P2 : Feature selection (8)

• P3 : Proximity-based learning (11)

• P4 : More linear models for prediction (12)

• P5 : Kernel methods (13)

• P6 : Tree-based methods and ensemble learning (14)

• Syllabus

• Chapter 1: Macroeconomic impact of the digital transition

• Chapter 2: Informational goods and replication costs

• Chapter 3: Consumers and transaction costs

• Chapter 4: Intermediation, matching and market design

• Chapter 5: Platform strategies

• Chapter 6: Data, pricing and auctions

• Chapter 7: Information and reputational effects

• Chapter 8: Recommender systems

• Chapter 9: Market concentration and dynamics

This course is an introduction to economic design via traditionnal information economics. 

Chapter 1 : Decision theory and microeconomic modelling

Chapter 2 : Non-strategic markets

Chapter 3 : Game theory

Chapter 4 : Oligopolies

Chapter 5 : Collusion

Chapter 6 : Innovation, entry and exists 

Chapter 1: From perfect competition to monopolies

Chapter 2: Price discrimination

Chapter 3: Oligopolies revisited

Chapter 4: Differenciation

Chapter 5: Vertical relationships