# Master Economics # Econometrics, big data, statistics

## RESPONSABLE

## AIM

The goal of this track is to train students to use methods of statistics and econometrics with a view to come up with relevant and solid answers to questions that businesses and administrations may face when making decisions. Beyond a solid knowledge of econometric methods and when they should be used, the students will be trained to apply them to real data and to present their results, in written or oral form, to various audiences. The students will be trained to use English in any professional context, i.e. to hold a conversation in English, to use technical vocabulary, to understand documents and articles, and to write in English.

## LINKS WITH RESEARCH

This Master’s degree is part of the *Ecole Universitaire de Recherche (EUR) AMSE*, which gathers together almost a hundred researchers from AMU, CNRS, EHESS and ECM.

The teachers are selected according to their expertise within those entities. The teaching staff is also supplemented with practitioners.

## FUNDAMENTAL PREREQUISITES

Two validated econometrics teachings.

## RECOMMENDED PREREQUISITES

Having followed teachings in statistics (estimating, testing, and confidence intervals) and econometrics of linear and nonlinear models. Teachings in statistical and econometric softwares and programming languages. This track is particularly adapted to students who have validated the first year of the Master of Economics in the Economics Department of AMSE, in the Faculty of Economics and Management at Aix-Marseille University. Access is possible in second year (M2).

## WEB SITE

## PLAQUETTE DE LA FORMATION

## PROFESSIONAL SKILLS

At the end of their second year (M2), every student will be competent in the main tools to manage and analyze the massive data and access a great part of the job offers for data analysts. Teaching is given partly in computer rooms in order to implement the tools taught. Pedagogy is based on the making of projects. The analytical mind of students is developed by an end-of-study internship involving the writing of a report. Professional skills targeted at the end of the second year :

- To determine the usefulness of statistical data in estimating the models likely to answer the question asked,
- to be able to manipulate and analyze quantitative and qualitative data whatever the size of the database,
- to choose relevant statistical and econometric tools and implement them so as to obtain reliable and robust answers likely to contribute to creating value for a business or public administration in the conduct of their actions,
- oral and written communication of the results from statistical and econometric analyses to various audiences.

## INTERNSHIPS AND SUPERVISED PROJECTS

At the end of the year, the students go through an internship and write a master’s internship report. The report aims at proving the student’s ability to use the conceptual tools acquired to questions pertaining to the professional world. The student must therefore identify the question, implement the tools, and be able to communicate the results to both a professional and academic audience. The internship is tutored by a scholar and an internship director (a member of the business). The report is defended in front of a jury comprised of the academic tutor, the internship direct, and two other people with the relevant skills (with at least a scholar).

## PLAQUETTE DE LA FORMATION

## EVALUATION AND EXAMS

Each course is assessed by a written exam or by an oral defense of a written portfolio. In order to limit the number of personal projects each student must put together, the teachers propose transversal projects unit when that is possible.

### Track EBDS (OPPT) (120 ECTS)

### M1 Economics (AN) (60 ECTS)

### S1 M1 ECO (SE) (30 ECTS)

### Microeconomics I and II (6 ECTS)

### Microeconomics I

## CONTENT

The objective of this course is to provide students with the foundations of economic theory. The course covers the consumption and production theory and is textbook based. The difficulty and coverage compare to those of the main departments of economics worldwide.

__Course outline :__

The course is textbook based. Topic list : Technology, Profit Maximization, Profit Function, Cost minimization, Cost Function, Duality, Utility Maximization, Choice, Demand.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Microeconomics II

## CONTENT

The objective of this course is to provide students with the foundations of economic theory. The course covers the consumption and production theory and is textbook based. The difficulty and coverage compare to those of the main departments of economics worldwide.

**Course outline :**

The course is textbook based. Topics list : Exchange, Time, Equilibrium Analysis, Welfare.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Macroeconomics I and II (6 ECTS)

### Macroeconomics I

## CONTENT

Learn the basic models with microeconomic foundations used in modern macroeconomics. Be able to do dynamic analysis. Understand the concept of dynamic efficiency and the role of public expenditures.

**Course outline :**

1. Introduction with reminders on the Solow model

2. The Ramsey model

2.1. The framework

2.2. Existence and features of the steady state

2.3. Dynamic analysis

2.4. Extension : public spending

3. The overlapping generations model

3.1. The model with capital

3.2. Intertemporal equilibrium, steady states and dynamics

3.3. Optimality

3.4. Extensions : public spending ; rational bubbles

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Macroeconomics II

## CONTENT

The aim of the course is to present advanced macroeconomic topics related to the analysis of aggregate consumption, aggregate investment and modern business cycle analysis with the Real Business Cycle model.

**Course outline :**

Chap. I : Consumption theory

1. Consumption over the life cycle : the life-cycle/permanent income models

2. Introducing uncertainty – The random walk hypothesis

3. Market imperfections : the role of liquidity constraints

4. Extensions : risk aversion, precautionary savings

Chap. 2 : Investment theory

1. The neoclassical model of capital demand

2. Investment with and without capital adjustment costs : Q-theory models

3. Role of shocks : real shocks, news shocks, noise shocks

Chap. 3 : Real Business Cycles

1. Measuring business cycles : trend-cycle decompositions and stylized facts

2. The canonical RBC model

3. Evaluation of the model

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Econometrics I and II (6 ECTS)

### Econometrics I: linear model

## CONTENT

**Provide students with :**

- the basics of panel data econometrics (fixed effects models, error components model)
- the identification of endogeneity problems in econometric models and their treatment (instrumental variables, GMM, tests)

**Course outline :**

1. Introduction to panel data and panel data models

2. The fixed effects model

- Specification of the model
- Estimation of the model : the Within / LSDV estimator.
- Testing the absence of unobserved heterogeneity.

3. The error components model

- Specification of the model
- Estimation of the model : the GLS / FGLS estimators.
- Testing the absence of unobserved heterogeneity.
- Testing the absence of correlation of the effects : the Hausman test

4. Endogeneity issues

- Causes of endogeneity in econometric models : measurement errors, dynamic models, unobserved heterogeneity, etc.
- The instrumental variables estimator
- The GMM estimator
- Looking for instruments (the time-series case, the cross-section case, the panel data case).
- Testing the validity of instruments
- Testing the exogeneity of regressors

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Econometrics II: non linear model

## CONTENT

Provide students with the basics of the econometrics of non linear models for binary, multinomial, ordered and count dependent variables as well as models for censored and truncated variables.

**Course outline :**

1. Introduction to non-linear models in econometrics and a brief reminder about the maximum likelihood principle.

2. Models for binary dependent variables.

- The linear probability model.
- The Logit model.
- The Probit model.

3. Models for multinomial and ordered dependent variables.

4. Models for count data.

5. Models for truncated and censored variables.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Labor economics - Risk and incentives (6 ECTS)

### Labor economics

## CONTENT

The objective of this course is to provide students with the necessary analytical tools to be able to study the consequences of different institutions, human capital formation, discrimination and wage bargaining on the labor market.

**Course outline :**

Introductory Chapter :

- Presentation
- Objectives
- Evaluation
- Labor market Institutions and course outline

Chapter 1 : « Labor Supply and Labor Demand » :

- Key definitions
- Labor Supply
- Labor Demand
- Equilibrium

Chapter 2. « Minimun wage »

- Facts
- Classical analysis
- The monopsony case
- Dual labor markets

Chapter 3 : « Mandatory contributions and social benefits »

- Facts
- Classical analysis
- Accounting for wage rigidities

Chapter 4. « Labor Unions » :

- Facts : unions, collective bargaining, union density
- The objective of labor unions
- Models of collective bargaining
- Model of strikes
- Empirical evidence and policy issues

Chapter 5. « Discrimination » :

- Facts : gender and ethnic wage and employment gaps
- Economic theories on discrimination
- Measuring wage discrimination
- Empirical results in the literature and policy issues

Chapter 6. « Education and human capital formation » :

- Facts
- The theory of human capital
- Education as a signaling device
- Identifying the causal relation between education and income
- Returns to education : private vs social returns

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Risk and incentives

## CONTENT

The main objective of this course is to provide the students with a theoretical synthetic framework so that they can face the difficulties of the study of economic decisions under uncertainty. Two main general topics will be dealt with : (1) the theory of decision under uncertainty, and (2) the moral hazard issues between several economic agents.

**Course outline :**

Chapter 1 : Risk, uncertainty and strategies

- Introduction of the main concepts (risk, uncertainty, probability, moral hazard, adverse selection)
- Probabilistic framework (space of states, random variables)
- Numerical decision criteria (preferences, representation by a numerical criteria)
- Game theory, Principal-Agent model

Chapter 2 : Expected Utility

- The virtues of the expected utility (Saint-Petersburg paradox)
- The axiomatics of the expected utility (objective and subjective expected utility)
- The limits of the expected utility (Allais paradox, Ellsberg paradox)
- Generalisations of the expected utility (rank-dependent expected utility, Choquet expected utility)

Chapter 3 : Risk Aversion and Risk Measures

- Qualitative approach (certainty equivalent, risk premium, risk attitude)
- Quantitative approach (local measures of risk aversion)
- Stochastic dominance (first and second order)

Chapter 4 : Introduction to moral hazard issues

- Risk sharing and sharecropping contracts
- Credit with risk aversion of the borrower

Chapter 5 : Other applications

- Risky saving
- Application of the expected utility to static portfolio choice

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Methodology I (6 ECTS)

### Software for economists I

## CONTENT

Provide students with the basics of the statistical and econometric treatment of data using SAS, from the statistical description of the sample, the detection of outliers to the implementation of estimation techniques for linear and non-linear models.

**Course outline :**

1. Introduction to SAS : importing and managing data – proc import, proc contents, proc format, proc sort, proc surveyselect and introduction to SAS macro functions.

2. Describing the data : descriptive statistics with SAS – proc means, proc univariate, proc freq, proc tabulate, proc gplot.

3. Estimating and testing linear models : proc reg, proc glm, proc model, proc panel.

4. Estimating and testing non linear models : proc logistic, proc probit, proc model, proc genmod, proc nlmixed.

## VOLUME OF TEACHINGS

- Tutorials:
**24**hours

### Mathematics for economists

## CONTENT

The course intends to deepen the understanding of optimization theory with a geometric approach, and to introduce in a second part the study of dynamical systems.

__Course outline__

**I. Optimization with mixed constraints** a. Tangent cone and KKT conditions b. Mixed constraints problem c. Constraints qualification conditions d. Convex problems e. Saddle point and duality

II. **Dynamical systems** a. Introduction b. Systems of linear equations

- Constant coefficient : resolution, exponantial of matrices
- Dynamic of the solutions : steady state, stability, planar systems
- nonhomogeneous systems c. Systems of nonlinear differential equations
- Existence and uniqueness theorem
- Linearized system, Hartman-Grobman theorem

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Elective course, choose one among two

### Refresher course in economics (0 ECTS)

## CONTENT

For students coming from other fields than economics : quick reminder about the fundamentals of Economics : utility and profit maximization, optimization, markets, equilibrium

**Course outline :**

1. Principles and key concepts of Economics

2. Foundations of Microeconomics : consumer decision and utility maximization

3. Foundations of Microeconomics : producer profit maximization and market equilibrium under perfect competition

## VOLUME OF TEACHINGS

- Lectures:
**6**hours

### Refresher course in mathematics and statistics (0 ECTS)

## CONTENT

For students who want to improve their math level : reminder about prerequisites for the Mathematics classes and basic notions of probability and statistics.

__Course outline :__

1. Linear algebra

2. Analysis and optimization

3. Matrix diagonalization

4. Ordinary differential equations of order 1

5. Basic notions of probability and statistics

## VOLUME OF TEACHINGS

- Lectures:
**6**hours

### S2 M1 ECO (SE) (30 ECTS)

### Microeconomics III and IV (6 ECTS)

### Microeconomics III - Game theory

## CONTENT

Introducing the basic concepts of Game Theory.

**Course outline :**

1. Complete information games (normal form, examples, analysis)

2. Mixed extension (lotteries, expected gain, mixed-strategy equilibrium)

3. Games with continuous actions (externalities,imperfect competition)

4. Incomplete information games (extensive form, subgame perfection)

5. Additional examples.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Microeconomics IV - Public economics

## CONTENT

The objective of this course is to study the role of state in the economy. It is designed to provide students with a broad overview of issues investigated in public economics. We will review the rational foundations of public intervention and explore some of the tools used by government to act : taxes and transfers, the provision of public goods, or the design of welfare schemes. Most topics will be approached from both theoretical and empirical points of view.

__Course outline :__

**Lecture 1 – Introduction to public economics**

- Foundations of public intervention – Normative and positive public economics – Some numbers about public intervention – Empirical methods for public economics

**Lecture 2 – Social choice and social welfare**

- Axiomatic approach to social choice – Social welfare functions

**Lecture 3 – Public goods and externalities**

- Public goods – Externalities

**Lecture 4 – Taxation of commodities**

- Tax incidence – Optimal commodity taxation

**Lecture 5 – Taxation of labor**

- Optimal labor taxation – Some empirics around labor taxation

**Lecture 6 – Taxation of capital**

- Taxes in an intertemporal framework – Optimal capital income taxation – Taxation of inheritances

**Lecture 7 – Social insurance**

- Unemployment insurance and workers’ compensation – Disability insurance – Health insurance

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Macroeconomics III and IV (6 ECTS)

### Macroeconomics III

## CONTENT

This course follows Macroeconomics II and it goes deeper in the description of micro-founded models by introducing market frictions into the RBC model. The DSGE-New Keynesian model which includes nominal rigidity is a natural extension of the RBC model to analyze monetary policy / fiscal policy. Although this course is mainly theoretical, lectures will be motivated by stylized facts and the empirical performance of business cycle models will be discussed.

**Course outline :**

Chapter 1 : Nominal rigidities (1) Introducing money in RBC model (2) Monopolistic competition (3) Price rigidity (4) Exercises

Chapter 2 : Monetary and Fiscal policy (1) Monetary policy analysis (2) Fiscal policy analysis (3) New topics in macroeconomics (ZLB, forward guidance…)

(4) Exercises

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Macroeconomics IV

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Methodology II (10 ECTS)

### Time series

## CONTENT

This course develops the basic theoretical tools for the analysis and estimation of univariate time series models. In particular, it discusses the concepts of stationarity and non-tationarity, unit-root tests, and exposes the techniques for estimating, forecasting and testing ARMA models using practical examples. Finally, it presents some non-linear models for conditional mean and variance.

**Course outline :**

• Brief Review of Statistics and Probability Concepts (pre-requisites)

• Stochastic processes and stationarity

• Classical stationary processes : AR, MA, ARMA

• Estimations techniques for the classical processes

• Forecasting methods for ARMA(p,q) processes

• White noise tests and stability tests

• Optimal choice of orders and Adequacy of parameters

• Univariate Non-Stationary processes and cointegration

• Modelling Nonlinearity of the conditional expectation

• Volatility modelling for univariate processes

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Software for economists II

## CONTENT

The objective of this course is twofold. Firstly, to study how to use and manipulate databases with Stata and secondly, to perform empirical analysis in relation with the concepts learned in the time series and econometric methods of evaluation classes. After a short introduction to Stata, the course will be divided into tasks-oriented sessions (with mini-projects and exercises) during which the students will perform empirical analysis using databases such as the World Values Survey, the French Labor Force Survey, the National Supported Work data, etc.

**Course outline :**

Lecture 1 : Introduction to Stata and database manipulation

Why using Stata – What Stata looks like – Importing and reading data into Stata – Examining the data – Saving the dataset – Keeping track of things – Organizing datasets – Creating new variables – Panel data manipulation

Lecture 2 : Graphs and linear regressions

Histograms – Two-dimensional graphs – Linear regressions – Post-estimation – Extracting results – Hypothesis testing – Interaction terms – Non-linearity – Fixed effects

Lecture 3 : Endogeneity and public policies econometrics

Randomized control trials – Difference-in-differences – Validity checks

Lecture 4 : Time series

Stationary and non-stationary processes

## VOLUME OF TEACHINGS

- Tutorials:
**24**hours

### Mathematics for finance

## CONTENT

**Objectives :**

Introducing elementary tools to analyse discrete and continuous-time random processes.

**Roadmap :**

1. Markov chains

1.1. Introductory Example : random walks

1.2. Markov chains on a finite state space

1.3. Markov chains on countable state spaces

1.3.1. States classification

1.3.2. Asymptotic results

2. Markovian processes in continuous time

2.1. Poisson processes

2.2. Continuous-time Markov processes

2.3. Queueing theory

3. Discrete-time random processes

3.1. Conditional expectation

3.2. Martingales

3.3. Stopping time

3.4. Convergence theorems

3.5. Applications

4. Introduction to continuous-time stochastic processes : Brownian motion

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Evaluation by econometric methods

## CONTENT

The objective of the course is to offer M1 students with an overview of the main empirical methods used for the evaluation of public policies. We will study key articles taken from various applied economics literature (health, education or activive labor policies). Practical case studies on STATA will be offered all along. We will point out advantages and limits of each method as well as guide in the selection of the appropriate method.

**Course outline :**

Introduction

1. Why evaluate ? What do we evaluate ? What is the objective ?

2. Potential outcome framework

3. Treatment effects and counterfactuals

4. Selection bias

Chapter 1 : Randomized experiments

1. Random assignment

2. Underlying assumptions

3. Study of 2 empirical papers using the method

4. Randomized experiments in practice

Running example : exercise on National Supported Work (NSW) data

Chapter 2 : Natural experiment : Difference-in-difference method

1. Model and underlying assumptions

2. Study of 2 empirical papers using the method

3. Extensions

4. D-in-D in practice

Running example : exercise on National Supported Work (NSW) data

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Elective teaching unit, choose 2 among 3 (8 ECTS)

### Project management - Health and environmental economics (4 ECTS)

### Project management

## CONTENT

Designing and managing development aid projects according to international standards.

**Course outline :**

The students will learn and practice how to build a development program.

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### Health and environmental economics

## CONTENT

The goal of this course is to bring together health and environmental economics as two narrow fields within the discipline of economics. This shall be done by identifying the interactions and intersections between health and environmental issues, describing the main economic properties that health and environment do have in common (market failure, externalities, government involvements …). This shall be followed by delineating the unique features of health and environment that make of them two distinct topics of study. Following this line of reasoning, the course shall, then, present two self-contained parts devoted to health economics and environmental economics. Each part shall present the workhorse analytical concepts and methods used by economists to explore specific issues relating to the two subfields. The course shall emphasize the use of economic evidence to identify priority issues and the most effective policies for health and environment. Examples and experiences of the kinds of topics that are addressed shall be provided all through the course.

**Course outline :**

Part I (4 hours) Overview : The links between health, environment and the economy

• Economic properties of health and environment

- What distinguishes “health goods” from “environmental goods” ?
- Typology of goods : Pure vs. impure public goods, private vs. publicly-provided private good, global vs. local public goods.

• The economic valuation approach :

- The theory of externalities
- Welfarism vs. extra-welfarism analysis
- Cost-benefit analysis, cost-effectiveness analysis, cost-utility analysis.
- Revealed vs. stated-preferences methods

• Government intervention and regulations

- Why do governments provide goods that are not pure public goods ? The case of health care services.
- What are the special characteristics and challenges of the “global public goods” ? The case of global climate change
- Rationing devices for publicly-provided goods (User charges, uniform provision, queuing).
- Efficiency conditions for public and pure public goods : Collective demand curve and provision of public goods.

Part II (7 hours) : Economics of Health and Health Care

• Overview : Health economics as a field of inquiry

• Health care market structure, conduct and performance :

- Do the law of supply and demand apply to health and health care ?
- What makes health and health care different ?
- Demand for health and health care : Health behavior
- Supply of health care : Production, provision and costs of health care.
- Health insurance markets : Public vs. private health insurance schemes, asymmetric information and agency, moral hazard and adverse selection.

• Reforming health care : Goals of reform, cost containment, efficiency and equity, extending insurance coverage, costs of universal coverage.

Part III (7 hours) : Economics of the Environment

• Overview : Economics and Environment :

- A framework of analysis, environmental microeconomics and macroeconomics.
- The environment as a public good.
- The global commons

• Ecological Economics and the Economic analysis of Environmental Issues :

- Valuing the environment, accounting for environmental costs, internalizing environmental costs, optimal pollution, the Coase theorem.
- Environmentally-adjusted national income accounts, the Genuine progress indicator, the better life index, environmental assets accounts.

• Environmental Health Policy : Impacts and Policy Responses :

- Measuring the economic cost of environmental impacts on health.
- Economic analysis and assessments of the performance of alternative policies in areas such as climate change, outdoor air pollution, water and sanitation.

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### Introduction to corporate finance - Financial econometrics (4 ECTS)

### Introduction to corporate finance

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### Financial econometrics

## CONTENT

1. Analyzing the properties of financial time series : application to French stock markets

The data consists of the stocks of the French CAC40 on a daily basis since 1980. Data are provided in excel format and need to be download to GRETL. Different companies are used as examples.

- Computing returns and historical volatility and analyzing their graphs (mean, variance, skewness and kurtosis, quantiles, min and max, autocorrelation)
- Analyzing the distributions of returns : non-parametric approaches (histograms and CDF based on kernels ; normality tests : QQ plot, Shapiro-Wilkinson, Doornik-Hansen, Jarque-Bera, etc.)
- Informal presentation of stable distributions : index of stability, skewness parameter, scale parameter, location parameter
- Example of parametrization of a stable distribution : the regression analysis of power law distributions.

2. Regression analysis of financial data

2.1.Evaluating the performance of a money manager : CAPM model

The data consist of the S&P 500 and some of its components (General Electric, Ford, Microsoft, ORACLE) and the 3-month Treasury bill).

- Estimate of the Betas using OLS and GLS
- Test of the CAPM using a two-pass regression
- The Jensen measure to evaluate manager performance.

2.2. Modellling the term structure of interest rates

The data consist of the Government zero-coupon bond yield taken at a daily frequency from 1990 to 2017 with several maturities : 6 months, 1 year, 2 years, 4, years, 4 years, 5 years, 7 years and 10 years.

- Analyzing some basic stylized facts of government bond yields (graphs of term to maturity, statistical properties, normality tets, correlation matrix, etc…).
- Recall on asset pricing, Duffie-Kan affine models and the decomposition of the yield curve.
- Decomposition of the tield curve using the Diebold’s regression approach : Level, slope and curvature curves.
- Factor models : a basic presentation of Kalman filter methodology and applications to the yield curve.

3.Some benchmark models for forecasting and trading models

The data consists of US/euro, US/Japan, US/UK exchange rate (daily) from 1999 and 1977 to 2017.

3.1.Models of naïve and MACD (moving average) strategies

3.2.ARMA models (identification via ACF and PACF, estimation, residual tests and forecasts)

3.3.Dectecting long-range dependence structure : an introduction to ARFIMA models

3.4.Introduction to stochastic volatility models : Harvey models and ARCH-GARCH models (tests and estimation).

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### Software for economists III - International trade (4 ECTS)

### Software for economists III

## CONTENT

This teaching unit aims at providing the fundamental basis of the use of R software (or the RStudio IDE) and R programming. The courses will be illustrated with exercises using the statistical environment R (http://www.r-project.org/) which is free, open-source free and multiplatform, or via the RStudio IDE. The organization of the course will make progressive the acquisition of the knowledge and the mastery of the R statistical tool. It aims to make the student more autonomous when faced to classical problem of statistical modelling or data analysis, which can be found in the fields of economics.

**Course outline :**

- Introduction (history).
- Basic handling (data management in R).
- Creating R functions.
- Loops, tests, vectorization.
- R Graphics.
- Application to modelling (regression/classification).

## VOLUME OF TEACHINGS

- Tutorials:
**18**hours

### International trade

## CONTENT

The aim of this course is to provide students the analytical tools that are essential to understand the causes and consequences of international trade. We will focus on some key questions as why nations trade, what they trade and who gains (or not) from trade. We will then analyse the reasons for countries to limit or regulate the exchange of goods and study the effects of such policies on development and inequality. We will also tackle some aspects of the globalization process like international norms, labor standards, firms’ organization, etc. We will heavily rely on formal economic modelling to help us understand issues of international trade.

**Course outline :**

1. Introduction – Basic facts

2. The Ricardian model

3. The Specific Factors model

4. The Hecksher-Ohlin Model

5. Trade theory with firm-level heterogeneity

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### M2 track EBDS (AN) (60 ECTS)

### S3 M2 EBDS (SE) (30 ECTS)

### Non-linear models: theory and applications (6 ECTS)

### Transition and duration models

## CONTENT

Students will study models of transitions and durations, and learn how to estimate these using real-world data.

__Course outline :__

This course is an introduction to modelling transitions into a state of interest (such as the transition into employment from unemployment) and durations (such as unemployment, survival of patients after medical treatment or firms after a financial crash). We start with the basic building blocks (Poisson processes, Markovian transitions, hazard models), and then develop methods for estimation using Maximum Likelihood. Duration data might be incomplete (hence are censored) in that we might not observe exits (individuals might still be in the state of interest at the end of the observation window), and unobserved heterogeneity introduces fundamental identification challenges.

Throughout all methods will be illustrated using examples in the software R, and we will consider several articles that have applied these methods. Several exercise sets will help students deepen their understanding of the theory.

**(I) Introduction to Poisson and counting processes**

The Poisson process is the classic counting process that models the arrival of new events, and thus transitions (increments) and durations (inter-arrival times). We will study this model, which has several interesting features, such as the independence of increments (transitions), so the Poisson process is a special Markov process (which exhibits a lack of memory property).

Application and illustrations in R : The number of doctor visits following a major health care reform (Poisson regressions). Search models of unemployment.

**(II) Introduction to Markov processes**

The Poisson process has increments that are independent, hence satisfy the Markov (lack of memory) property. We generalise this idea, consider transition probabilities between states, and look at the evolution of the Markov process in time.

Applications and illustrations : we will study numerical examples using R, and several applications such as Nakajima (2007, ReStud), “Measuring Peer Effects on Youth Smoking Behaviour”, and Topa (2001, ReStud), “Social Interactions, Local Spillovers and Unemployment.”

**(III) Duration and survival analysis : Hazard models**

The Poisson model gives a simple duration model, but in many empirical situations this is too limiting as exit rates (hazards) form the state of interest are constant. In many situations the exit rate depends on the duration of stay, for instance the longer the unemployment spell the less likely the exit (duration dependence). We will study the main modelling objects (hazard rate and survival function), examine several parametric models (e.g. the Weibull model), and incorporate observed heterogeneity across individuals (Cox’s proportional hazard (PH) model). In order to accommodate unmeasured heterogeneity, we extend the PH model to the mixed proportional hazard (MPH) model. This unobserved heterogeneity will introduce a fundamental identification problem since duration dependence might be confounded by dynamic sorting (individuals of a latent “high” type tending to exit the state of interest more quickly). Since the models are fully parametrically specified, it is natural to estimate these using Maximum Likelihood. One particular characteristic of duration data is that we might not observe an individual’s exit from the state of interest ; hence these duration data might be incomplete (hence are censored), and this needs to be taken into account in the estimation.

Applications in R : Survival probabilities for smokers, criminal recidivism

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Models for truncated and censored variables

## CONTENT

The main objective of this course is to provide the students with a synthetic framework so that they can thoroughly understand and efficiently apply to concrete cases the main estimation and test techniques for limited-dependent variable models with censorship or truncation.

**Course outline :**

1. Brief remainder of the foundations of the specification and estimation of econometric models (GMM, Likelihood maximisation, Simulations)

2. Presentation of censored and truncated data, limited-dependent variables and their distributions

3. Selection bias, efficiency loss, punctual and partial identification

4. Models with censored or truncated dependent variables

5. Roy model and extensions

6. Panel data models with unobserved heterogeneity or compound errors

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Advanced econometrics I: theory and applications (6 ECTS)

### Non parametric methods in econometrics

## CONTENT

Non Parametric methods are statistical techniques that do not require to specify functional forms for objects being estimated. Instead, they let the data itself plays and informs the resulting model in a particular manner. Such methods are becoming increasingly popular for applied data analysis, they are best suited to situations involving large data sets for which the number of variables involved is manageable. These methods are often deployed after common parametric specifications are found to be unsuitable for the problem at hand, particularly when formal rejection of a parametric model based on specification tests yields no clues as to the direction in which to search for an improved parametric model.

The job market understood the importance of the non/semi-parametric methods and almost any serious software contains the principal techniques in this area. We illustrate the different models and techniques with R and Matlab. First, R because of the huge number of packages from CRAN, and secondly Matlab because is the easiest environment for programming arrays in econometrics (and typically all objects are arrays in applied econometrics). Both are very representative for the job market. Each lecture will be accompanied by numerical examples and small programming tutorials.

**Course outline :**

- Brief Review of Statistics and Probability Concepts used in the course (pre-requisites)
- Empirical non parametric estimators : histogram, empirical distribution, regressogram
- Kernel-based methods in Non Parametric Econometrics : Parzen-Rosenblatt, Nadaraya-Watson
- Semi-parametric methods : Single index and Additive models
- Splines-based methods in Non Parametric Econometrics : Smoothing splines and Least-Squares splines
- Series-based methods in Non Parametric Econometrics : Wavelets, Laguerre and Hermite Polynomials

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Multivariate and non-linear time series

## CONTENT

Make the students able to cope with multivariate, nonlinear time series analysis.

__Course outline :__

- Introduction to time series analysis
- Nonlinear models (TAR, STAR, MS, tests, forecasting
- Multivariate models (VAR, SVAR, cointegration
- Factor analysis (Macro, fundamental, PCA)
- Multivariate volatility models (VEC, BEKK, DCC, OGARCH)

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Advanced econometrics II: theory and applications (6 ECTS)

### Methodology of econometrics and statistical studies

## CONTENT

Provide students with a set of rules to be followed in the course of realization of a statistical or econometric study for an organization (be it a private company, a public or private administration, a NGO, etc.).

Course outline :

- Starting from the beginning : calls for tender, mutual agreement contracts and within organization projects.
- Identifying the question(s) to be answered. Modelling issues.
- Going from the economic model to the econometric model.
- Collecting, inspecting and managing the data.
- Choosing the right statistical or econometric estimation method(s).
- Analysing the results and their consequences in terms of decision/policy. Professional ethics.
- Communicating about the results in oral and written forms.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Advanced econometrics

## CONTENT

The goal of this course is to present advanced methods in econometrics for distributional analysis, regression and classification models. The course will present theoretical foundations and underlying intuition of each method, as well as several empirical examples.

**Course outline :**

1. Resampling Methods

- Pseudo-random generator
- Monte Carlo experiments
- Bootstrap and permutation tests

2. Nonparametric Econometrics

- Density estimation
- Regression splines
- Finite mixture models

3. Econometrics and Machine Learning

- Philosophy and general principle
- Resampling-based methods and algorithms
- Misspecification detection

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Languages, softwares and tools for Big Data (6 ECTS)

### Programming for Big Data (Python, SQL, noSQL, etc)

## CONTENT

This course is aimed at teaching the basics of computer programming, with emphasis on its use in Big Data. Students will first become familiar with database management. They will then learn the basics of programming with the computer language Python.

**Course outline :**

1. Database Management (relational model, relational algebra, SQL language, …)

2. Introduction to Python

3. Creating Functions

4. Introduction to Numpy

5. Data manipulation with Pandas

6. Visualization

7. Parallel programming

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Logiciels pour les big data

*Unavailable contents.*

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Applications for Big Data: elective teaching units, choose 2 among 4 (6 ECTS)

### Big data and public policies (3 ECTS)

### Big data and public policies

## CONTENT

The objective is to introduce several questions associated with the use of big data to understand the context, impacts and limits of this new technology. Issues will be presented to students : e.g. access, security, innovation and opportunities to specifically identify how big data can boost local development and allow to design new public policies.

**Course outline :**

Part I : Data and local development

I. Open data, Smart Region and innovation

II. Contracts and legal aspect of the platform FlexGrid, a tool for energy transition

Part II : Big data and security

I. Public security and cities security

II. Cybersecurity

Part II : Big data, a decision-making support

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big data and quantitative marketing (3 ECTS)

### Big data and quantitative marketing

## CONTENT

Understand how data analytics, machine and deep learning methods on large volumes of structured or unstructured data allow to better model, predict or describe consumer behaviour. Understand how data analytics based modelling impacts marketing decisions and how recent advances are changing market research and marketing science.

**Course outline :**

- The 3 main uses of data in marketing analytics :

o Model and describe behaviours / segment consumers o Predict behaviours based on a set of variables o Receive real-time information on consumer behaviours

- The types of data used in quantitative marketing :

o Aggregate statistical/econometric data and time series o Individual observations in data sets (survey data, databases, etc.) o User generated content (text and images) o Datafied objects

- How data is changing and why : from solicited to unsolicited, from structured to unstructured, from manual datafication to algorithm-led datafication, from text to data, etc.
- The main statistical methods used in data analytics
- The code languages used in advanced data analytics
- Algorithms, machine learning, deep learning and artificial intelligence

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big data and finance (3 ECTS)

### Big data and finance

## CONTENT

The course presents the last developments around the use of big data technics in finance. The first part offers an overview of the various recent applications to corporate finance and financial regulation. The second concentrates on the use of big data and associated models in market finance. The third and last part highlights the role of these methods in insurance and reinsurance market.

**Course outline :**

Part 1 Overview of the applications of Big data in finance (6h, Pierre Bittner)

1– The interest of big data in finance

1.1 – Reminder on Big data

1.2 – Big data and decision

1.3 – Big data and market supervision

2– Case study in finance

2.1 – Applications to corporate and investment banking

2.2 – Regulatory challenges

Part 2 : Big data and market finance (12h, Yoann Bourgeois)

1- Realized Volatility

1.1- Continuous time pricing fundamentals

1.1.1 Brownian motion and random walk

1.1.2 Stochastic Differential Equation/ Stochastic Integrals

1.1.3 Quadratic Variation

1.1.4 Implied volatility in Black Scholes

1.2- Realized Volatility

1.2.1 Unbiased estimators

1.2.3 Confidence intervals

1.2.3 Application FX market

1.3-RV and integrated variance

1.3.1 Seasonality

1.3.2 The impact of periodic events on the RV.

1.3.3 Application FX market

2- Bonds portfolio automatic engine

2.1 Definitions (Yields, Bond, Duration, P&L of a bond etc.)

2.2 Bonds clustering (PCA+KMeans)

2.3 The reference curve construction

2.3.1 Regression

2.3.2 Cubic Splines

2.4 Z-Score and momentum to sort bonds

2.5 Reference bonds replication

2.6 Application France 10Y reference bond.

3- Intraday hedging of FX options

3.1 SABR model

3.2 Gatheral parametric local volatility model

3.3 Intraday model calibrations

3.4 Tichonov Regularization

3.5 The use of risk neutral distribution quantiles and moments

3.6 Application FX vanilla options

Part 3 : Big data and insurance (6h, Serdar Coskun)

This part presents, via recent cases, the utility of big data and insurance and reinsurance markets. It also presents the most recent progress in the « insurtech » market.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big data: other applications (3 ECTS)

### Big data: other applications

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### S4 M2 EBDS (SE) (30 ECTS)

### Advanced methods in Big Data (9 ECTS)

### Automatic model selection methods

## CONTENT

The objective of this course is to introduce quantitative methods allowing to reduce information. These methods cover different fields of statistics and are based on classical econometric methods (OLS, MLE) or classificatory or principal component methods. The ultimate goal is to study methods to do automatic variable selection in large-scale problems and to apply them to real data.

**Course outline :**

- Classification methods
- Economic factor models
- Statistical factor models
- Lasso methods
- The so-called « General to Specific » method (Hendry, Gets or Autometrics Methodology)

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Predictive methods

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Machine learning and statistical learning

## CONTENT

This course provides a broad introduction to statistical learning and machine learning. The main objective is to provide students with the knowledge necessary to understand machine learning methods.

**Course outline :**

Introduction : Statistical learning and machine learning : what and why ?

Part 1. Unsupervised learning

Clustering

Expectation maximization

Dimensionality reduction

Part 2. Supervised learning : regression

Linear regression

Logistic regression

Part 3. Supervised learning : machine learning

Generative models

Bias/variance tradeoff

Cross-validation

Variable selection

Part 4. Machine learning algorithms by hand

Classification and regression trees, Bagging, Random Forest, Boosting, Support Vector Machine, Neural Networks, etc.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### End-of-study internship with report and defence (21 ECTS)

### Track EBDS magistère option (OPPT) (144 ECTS)

### M1 Economics (magistère option) (AN) (72 ECTS)

### S1 M1 Economics magistère option (SE) (36 ECTS)

### Microeconomics I and II (6 ECTS)

### Microeconomics I

## CONTENT

The objective of this course is to provide students with the foundations of economic theory. The course covers the consumption and production theory and is textbook based. The difficulty and coverage compare to those of the main departments of economics worldwide.

__Course outline :__

The course is textbook based. Topic list : Technology, Profit Maximization, Profit Function, Cost minimization, Cost Function, Duality, Utility Maximization, Choice, Demand.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Microeconomics II

## CONTENT

**Course outline :**

The course is textbook based. Topics list : Exchange, Time, Equilibrium Analysis, Welfare.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Macroeconomics I and II (6 ECTS)

### Macroeconomics I

## CONTENT

Learn the basic models with microeconomic foundations used in modern macroeconomics. Be able to do dynamic analysis. Understand the concept of dynamic efficiency and the role of public expenditures.

**Course outline :**

1. Introduction with reminders on the Solow model

2. The Ramsey model

2.1. The framework

2.2. Existence and features of the steady state

2.3. Dynamic analysis

2.4. Extension : public spending

3. The overlapping generations model

3.1. The model with capital

3.2. Intertemporal equilibrium, steady states and dynamics

3.3. Optimality

3.4. Extensions : public spending ; rational bubbles

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Macroeconomics II

## CONTENT

The aim of the course is to present advanced macroeconomic topics related to the analysis of aggregate consumption, aggregate investment and modern business cycle analysis with the Real Business Cycle model.

**Course outline :**

Chap. I : Consumption theory

1. Consumption over the life cycle : the life-cycle/permanent income models

2. Introducing uncertainty – The random walk hypothesis

3. Market imperfections : the role of liquidity constraints

4. Extensions : risk aversion, precautionary savings

Chap. 2 : Investment theory

1. The neoclassical model of capital demand

2. Investment with and without capital adjustment costs : Q-theory models

3. Role of shocks : real shocks, news shocks, noise shocks

Chap. 3 : Real Business Cycles

1. Measuring business cycles : trend-cycle decompositions and stylized facts

2. The canonical RBC model

3. Evaluation of the model

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Econometrics I and II (6 ECTS)

### Econometrics I: linear model

## CONTENT

**Provide students with :**

- the basics of panel data econometrics (fixed effects models, error components model)
- the identification of endogeneity problems in econometric models and their treatment (instrumental variables, GMM, tests)

**Course outline :**

1. Introduction to panel data and panel data models

2. The fixed effects model

- Specification of the model
- Estimation of the model : the Within / LSDV estimator.
- Testing the absence of unobserved heterogeneity.

3. The error components model

- Specification of the model
- Estimation of the model : the GLS / FGLS estimators.
- Testing the absence of unobserved heterogeneity.
- Testing the absence of correlation of the effects : the Hausman test

4. Endogeneity issues

- Causes of endogeneity in econometric models : measurement errors, dynamic models, unobserved heterogeneity, etc.
- The instrumental variables estimator
- The GMM estimator
- Looking for instruments (the time-series case, the cross-section case, the panel data case).
- Testing the validity of instruments
- Testing the exogeneity of regressors

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Econometrics II: non linear model

## CONTENT

Provide students with the basics of the econometrics of non linear models for binary, multinomial, ordered and count dependent variables as well as models for censored and truncated variables.

**Course outline :**

1. Introduction to non-linear models in econometrics and a brief reminder about the maximum likelihood principle.

2. Models for binary dependent variables.

- The linear probability model.
- The Logit model.
- The Probit model.

3. Models for multinomial and ordered dependent variables.

4. Models for count data.

5. Models for truncated and censored variables.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Labor economics - Risk and incentives (6 ECTS)

### Labor economics

## CONTENT

The objective of this course is to provide students with the necessary analytical tools to be able to study the consequences of different institutions, human capital formation, discrimination and wage bargaining on the labor market.

**Course outline :**

Introductory Chapter :

- Presentation
- Objectives
- Evaluation
- Labor market Institutions and course outline

Chapter 1 : « Labor Supply and Labor Demand » :

- Key definitions
- Labor Supply
- Labor Demand
- Equilibrium

Chapter 2. « Minimun wage »

- Facts
- Classical analysis
- The monopsony case
- Dual labor markets

Chapter 3 : « Mandatory contributions and social benefits »

- Facts
- Classical analysis
- Accounting for wage rigidities

Chapter 4. « Labor Unions » :

- Facts : unions, collective bargaining, union density
- The objective of labor unions
- Models of collective bargaining
- Model of strikes
- Empirical evidence and policy issues

Chapter 5. « Discrimination » :

- Facts : gender and ethnic wage and employment gaps
- Economic theories on discrimination
- Measuring wage discrimination
- Empirical results in the literature and policy issues

Chapter 6. « Education and human capital formation » :

- Facts
- The theory of human capital
- Education as a signaling device
- Identifying the causal relation between education and income
- Returns to education : private vs social returns

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Risk and incentives

## CONTENT

The main objective of this course is to provide the students with a theoretical synthetic framework so that they can face the difficulties of the study of economic decisions under uncertainty. Two main general topics will be dealt with : (1) the theory of decision under uncertainty, and (2) the moral hazard issues between several economic agents.

**Course outline :**

Chapter 1 : Risk, uncertainty and strategies

- Introduction of the main concepts (risk, uncertainty, probability, moral hazard, adverse selection)
- Probabilistic framework (space of states, random variables)
- Numerical decision criteria (preferences, representation by a numerical criteria)
- Game theory, Principal-Agent model

Chapter 2 : Expected Utility

- The virtues of the expected utility (Saint-Petersburg paradox)
- The axiomatics of the expected utility (objective and subjective expected utility)
- The limits of the expected utility (Allais paradox, Ellsberg paradox)
- Generalisations of the expected utility (rank-dependent expected utility, Choquet expected utility)

Chapter 3 : Risk Aversion and Risk Measures

- Qualitative approach (certainty equivalent, risk premium, risk attitude)
- Quantitative approach (local measures of risk aversion)
- Stochastic dominance (first and second order)

Chapter 4 : Introduction to moral hazard issues

- Risk sharing and sharecropping contracts
- Credit with risk aversion of the borrower

Chapter 5 : Other applications

- Risky saving
- Application of the expected utility to static portfolio choice

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Methodology I (6 ECTS)

### Software for economists I

## CONTENT

Provide students with the basics of the statistical and econometric treatment of data using SAS, from the statistical description of the sample, the detection of outliers to the implementation of estimation techniques for linear and non-linear models.

**Course outline :**

1. Introduction to SAS : importing and managing data – proc import, proc contents, proc format, proc sort, proc surveyselect and introduction to SAS macro functions.

2. Describing the data : descriptive statistics with SAS – proc means, proc univariate, proc freq, proc tabulate, proc gplot.

3. Estimating and testing linear models : proc reg, proc glm, proc model, proc panel.

4. Estimating and testing non linear models : proc logistic, proc probit, proc model, proc genmod, proc nlmixed.

## VOLUME OF TEACHINGS

- Tutorials:
**24**hours

### Mathematics for economists

## CONTENT

The course intends to deepen the understanding of optimization theory with a geometric approach, and to introduce in a second part the study of dynamical systems.

__Course outline__

**I. Optimization with mixed constraints** a. Tangent cone and KKT conditions b. Mixed constraints problem c. Constraints qualification conditions d. Convex problems e. Saddle point and duality

II. **Dynamical systems** a. Introduction b. Systems of linear equations

- Constant coefficient : resolution, exponantial of matrices
- Dynamic of the solutions : steady state, stability, planar systems
- nonhomogeneous systems c. Systems of nonlinear differential equations
- Existence and uniqueness theorem
- Linearized system, Hartman-Grobman theorem

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big Data (6 ECTS)

### Big data, challenges and opportunities

## VOLUME OF TEACHINGS

- Lectures:
**6**hours

### Programming for Big Data, an introduction to Python and SQL

## CONTENT

This course is aimed at teaching the basics of computer programming, with emphasis on its use in Big Data. Students will first become familiar with database management (relational or not). They will then learn the basics of programming with the computer language Python.

**Course outline :**

Chapter 1 : Relational databases

1. Introduction

2. The relational model

3. Relational algebra

4. SQL Language

5. Entity-Association Schemes

Chapter 2 : Non-relational databases

1. Introduction

2. Parallel Computing

3. Schemas, and non-relational databases

4. MongoDB

Chapter 3 : Introduction to Python

1. Variables et calculs

2. Strings, lists, tuples, dictionnaries

3. If… else conditions

4. Loops

5. Functions

6. Introduction to Numpy

7. Data handling with Pandas

8. Visualization

9. Parallel programming

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big data softwares

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### S2 M1 Economics magistère option (SE) (36 ECTS)

### Microeconomics III and IV (6 ECTS)

### Microeconomics III - Game theory

## CONTENT

Introducing the basic concepts of Game Theory.

**Course outline :**

1. Complete information games (normal form, examples, analysis)

2. Mixed extension (lotteries, expected gain, mixed-strategy equilibrium)

3. Games with continuous actions (externalities,imperfect competition)

4. Incomplete information games (extensive form, subgame perfection)

5. Additional examples.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Microeconomics IV - Public economics

## CONTENT

The objective of this course is to study the role of state in the economy. It is designed to provide students with a broad overview of issues investigated in public economics. We will review the rational foundations of public intervention and explore some of the tools used by government to act : taxes and transfers, the provision of public goods, or the design of welfare schemes. Most topics will be approached from both theoretical and empirical points of view.

__Course outline :__

**Lecture 1 – Introduction to public economics**

- Foundations of public intervention – Normative and positive public economics – Some numbers about public intervention – Empirical methods for public economics

**Lecture 2 – Social choice and social welfare**

- Axiomatic approach to social choice – Social welfare functions

**Lecture 3 – Public goods and externalities**

- Public goods – Externalities

**Lecture 4 – Taxation of commodities**

- Tax incidence – Optimal commodity taxation

**Lecture 5 – Taxation of labor**

- Optimal labor taxation – Some empirics around labor taxation

**Lecture 6 – Taxation of capital**

- Taxes in an intertemporal framework – Optimal capital income taxation – Taxation of inheritances

**Lecture 7 – Social insurance**

- Unemployment insurance and workers’ compensation – Disability insurance – Health insurance

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big Data (6 ECTS)

### Advanced SAS

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Introduction to machine learning

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Macroeconomics III and IV (6 ECTS)

### Macroeconomics III

## CONTENT

This course follows Macroeconomics II and it goes deeper in the description of micro-founded models by introducing market frictions into the RBC model. The DSGE-New Keynesian model which includes nominal rigidity is a natural extension of the RBC model to analyze monetary policy / fiscal policy. Although this course is mainly theoretical, lectures will be motivated by stylized facts and the empirical performance of business cycle models will be discussed.

**Course outline :**

Chapter 1 : Nominal rigidities (1) Introducing money in RBC model (2) Monopolistic competition (3) Price rigidity (4) Exercises

Chapter 2 : Monetary and Fiscal policy (1) Monetary policy analysis (2) Fiscal policy analysis (3) New topics in macroeconomics (ZLB, forward guidance…)

(4) Exercises

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Macroeconomics IV

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Méthodologie (6 ECTS)

### Software for economists II

## CONTENT

The objective of this course is twofold. Firstly, to study how to use and manipulate databases with Stata and secondly, to perform empirical analysis in relation with the concepts learned in the time series and econometric methods of evaluation classes. After a short introduction to Stata, the course will be divided into tasks-oriented sessions (with mini-projects and exercises) during which the students will perform empirical analysis using databases such as the World Values Survey, the French Labor Force Survey, the National Supported Work data, etc.

**Course outline :**

Lecture 1 : Introduction to Stata and database manipulation

Why using Stata – What Stata looks like – Importing and reading data into Stata – Examining the data – Saving the dataset – Keeping track of things – Organizing datasets – Creating new variables – Panel data manipulation

Lecture 2 : Graphs and linear regressions

Histograms – Two-dimensional graphs – Linear regressions – Post-estimation – Extracting results – Hypothesis testing – Interaction terms – Non-linearity – Fixed effects

Lecture 3 : Endogeneity and public policies econometrics

Randomized control trials – Difference-in-differences – Validity checks

Lecture 4 : Time series

Stationary and non-stationary processes

## VOLUME OF TEACHINGS

- Tutorials:
**24**hours

### Mathematics for finance

## CONTENT

**Objectives :**

Introducing elementary tools to analyse discrete and continuous-time random processes.

**Roadmap :**

1. Markov chains

1.1. Introductory Example : random walks

1.2. Markov chains on a finite state space

1.3. Markov chains on countable state spaces

1.3.1. States classification

1.3.2. Asymptotic results

2. Markovian processes in continuous time

2.1. Poisson processes

2.2. Continuous-time Markov processes

2.3. Queueing theory

3. Discrete-time random processes

3.1. Conditional expectation

3.2. Martingales

3.3. Stopping time

3.4. Convergence theorems

3.5. Applications

4. Introduction to continuous-time stochastic processes : Brownian motion

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Vocational courses (6 ECTS)

### Quantitative marketing

## VOLUME OF TEACHINGS

- Lectures:
**12**hours

### Software: R

## VOLUME OF TEACHINGS

- Lectures:
**12**hours

### Economic policy II

## VOLUME OF TEACHINGS

- Lectures:
**12**hours

### Insurance mechanisms

## VOLUME OF TEACHINGS

- Lectures:
**12**hours

### Oral training on Economics topics

## VOLUME OF TEACHINGS

- Tutorials:
**6**hours

### Oral training in English

## VOLUME OF TEACHINGS

- Tutorials:
**6**hours

### Elective courses, choose 2 among 6 (6 ECTS)

### Introduction to corporate finance (3 ECTS)

### Introduction to corporate finance

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### Project management (3 ECTS)

### Project management

## CONTENT

Designing and managing development aid projects according to international standards.

**Course outline :**

The students will learn and practice how to build a development program.

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### Health and environmental economics (3 ECTS)

### Health and environmental economics

## CONTENT

The goal of this course is to bring together health and environmental economics as two narrow fields within the discipline of economics. This shall be done by identifying the interactions and intersections between health and environmental issues, describing the main economic properties that health and environment do have in common (market failure, externalities, government involvements …). This shall be followed by delineating the unique features of health and environment that make of them two distinct topics of study. Following this line of reasoning, the course shall, then, present two self-contained parts devoted to health economics and environmental economics. Each part shall present the workhorse analytical concepts and methods used by economists to explore specific issues relating to the two subfields. The course shall emphasize the use of economic evidence to identify priority issues and the most effective policies for health and environment. Examples and experiences of the kinds of topics that are addressed shall be provided all through the course.

**Course outline :**

Part I (4 hours) Overview : The links between health, environment and the economy

• Economic properties of health and environment

- What distinguishes “health goods” from “environmental goods” ?
- Typology of goods : Pure vs. impure public goods, private vs. publicly-provided private good, global vs. local public goods.

• The economic valuation approach :

- The theory of externalities
- Welfarism vs. extra-welfarism analysis
- Cost-benefit analysis, cost-effectiveness analysis, cost-utility analysis.
- Revealed vs. stated-preferences methods

• Government intervention and regulations

- Why do governments provide goods that are not pure public goods ? The case of health care services.
- What are the special characteristics and challenges of the “global public goods” ? The case of global climate change
- Rationing devices for publicly-provided goods (User charges, uniform provision, queuing).
- Efficiency conditions for public and pure public goods : Collective demand curve and provision of public goods.

Part II (7 hours) : Economics of Health and Health Care

• Overview : Health economics as a field of inquiry

• Health care market structure, conduct and performance :

- Do the law of supply and demand apply to health and health care ?
- What makes health and health care different ?
- Demand for health and health care : Health behavior
- Supply of health care : Production, provision and costs of health care.
- Health insurance markets : Public vs. private health insurance schemes, asymmetric information and agency, moral hazard and adverse selection.

• Reforming health care : Goals of reform, cost containment, efficiency and equity, extending insurance coverage, costs of universal coverage.

Part III (7 hours) : Economics of the Environment

• Overview : Economics and Environment :

- A framework of analysis, environmental microeconomics and macroeconomics.
- The environment as a public good.
- The global commons

• Ecological Economics and the Economic analysis of Environmental Issues :

- Valuing the environment, accounting for environmental costs, internalizing environmental costs, optimal pollution, the Coase theorem.
- Environmentally-adjusted national income accounts, the Genuine progress indicator, the better life index, environmental assets accounts.

• Environmental Health Policy : Impacts and Policy Responses :

- Measuring the economic cost of environmental impacts on health.
- Economic analysis and assessments of the performance of alternative policies in areas such as climate change, outdoor air pollution, water and sanitation.

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### Evaluation by econometric methods (3 ECTS)

### Evaluation by econometric methods

## CONTENT

The objective of the course is to offer M1 students with an overview of the main empirical methods used for the evaluation of public policies. We will study key articles taken from various applied economics literature (health, education or activive labor policies). Practical case studies on STATA will be offered all along. We will point out advantages and limits of each method as well as guide in the selection of the appropriate method.

**Course outline :**

Introduction

1. Why evaluate ? What do we evaluate ? What is the objective ?

2. Potential outcome framework

3. Treatment effects and counterfactuals

4. Selection bias

Chapter 1 : Randomized experiments

1. Random assignment

2. Underlying assumptions

3. Study of 2 empirical papers using the method

4. Randomized experiments in practice

Running example : exercise on National Supported Work (NSW) data

Chapter 2 : Natural experiment : Difference-in-difference method

1. Model and underlying assumptions

2. Study of 2 empirical papers using the method

3. Extensions

4. D-in-D in practice

Running example : exercise on National Supported Work (NSW) data

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### International trade (3 ECTS)

### International trade

## CONTENT

The aim of this course is to provide students the analytical tools that are essential to understand the causes and consequences of international trade. We will focus on some key questions as why nations trade, what they trade and who gains (or not) from trade. We will then analyse the reasons for countries to limit or regulate the exchange of goods and study the effects of such policies on development and inequality. We will also tackle some aspects of the globalization process like international norms, labor standards, firms’ organization, etc. We will heavily rely on formal economic modelling to help us understand issues of international trade.

**Course outline :**

1. Introduction – Basic facts

2. The Ricardian model

3. The Specific Factors model

4. The Hecksher-Ohlin Model

5. Trade theory with firm-level heterogeneity

## VOLUME OF TEACHINGS

- Lectures:
**18**hours

### M2 track EBDS magistère option (AN) (72 ECTS)

### S3 M2 EBDS magistère option (SE) (36 ECTS)

### Non-linear models: theory and applications (6 ECTS)

### Transition and duration models

## CONTENT

Students will study models of transitions and durations, and learn how to estimate these using real-world data.

__Course outline :__

This course is an introduction to modelling transitions into a state of interest (such as the transition into employment from unemployment) and durations (such as unemployment, survival of patients after medical treatment or firms after a financial crash). We start with the basic building blocks (Poisson processes, Markovian transitions, hazard models), and then develop methods for estimation using Maximum Likelihood. Duration data might be incomplete (hence are censored) in that we might not observe exits (individuals might still be in the state of interest at the end of the observation window), and unobserved heterogeneity introduces fundamental identification challenges.

Throughout all methods will be illustrated using examples in the software R, and we will consider several articles that have applied these methods. Several exercise sets will help students deepen their understanding of the theory.

**(I) Introduction to Poisson and counting processes**

The Poisson process is the classic counting process that models the arrival of new events, and thus transitions (increments) and durations (inter-arrival times). We will study this model, which has several interesting features, such as the independence of increments (transitions), so the Poisson process is a special Markov process (which exhibits a lack of memory property).

Application and illustrations in R : The number of doctor visits following a major health care reform (Poisson regressions). Search models of unemployment.

**(II) Introduction to Markov processes**

The Poisson process has increments that are independent, hence satisfy the Markov (lack of memory) property. We generalise this idea, consider transition probabilities between states, and look at the evolution of the Markov process in time.

Applications and illustrations : we will study numerical examples using R, and several applications such as Nakajima (2007, ReStud), “Measuring Peer Effects on Youth Smoking Behaviour”, and Topa (2001, ReStud), “Social Interactions, Local Spillovers and Unemployment.”

**(III) Duration and survival analysis : Hazard models**

The Poisson model gives a simple duration model, but in many empirical situations this is too limiting as exit rates (hazards) form the state of interest are constant. In many situations the exit rate depends on the duration of stay, for instance the longer the unemployment spell the less likely the exit (duration dependence). We will study the main modelling objects (hazard rate and survival function), examine several parametric models (e.g. the Weibull model), and incorporate observed heterogeneity across individuals (Cox’s proportional hazard (PH) model). In order to accommodate unmeasured heterogeneity, we extend the PH model to the mixed proportional hazard (MPH) model. This unobserved heterogeneity will introduce a fundamental identification problem since duration dependence might be confounded by dynamic sorting (individuals of a latent “high” type tending to exit the state of interest more quickly). Since the models are fully parametrically specified, it is natural to estimate these using Maximum Likelihood. One particular characteristic of duration data is that we might not observe an individual’s exit from the state of interest ; hence these duration data might be incomplete (hence are censored), and this needs to be taken into account in the estimation.

Applications in R : Survival probabilities for smokers, criminal recidivism

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Models for truncated and censored variables

## CONTENT

The main objective of this course is to provide the students with a synthetic framework so that they can thoroughly understand and efficiently apply to concrete cases the main estimation and test techniques for limited-dependent variable models with censorship or truncation.

**Course outline :**

1. Brief remainder of the foundations of the specification and estimation of econometric models (GMM, Likelihood maximisation, Simulations)

2. Presentation of censored and truncated data, limited-dependent variables and their distributions

3. Selection bias, efficiency loss, punctual and partial identification

4. Models with censored or truncated dependent variables

5. Roy model and extensions

6. Panel data models with unobserved heterogeneity or compound errors

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Advanced econometrics I: theory and applications (6 ECTS)

### Non parametric methods in econometrics

## CONTENT

Non Parametric methods are statistical techniques that do not require to specify functional forms for objects being estimated. Instead, they let the data itself plays and informs the resulting model in a particular manner. Such methods are becoming increasingly popular for applied data analysis, they are best suited to situations involving large data sets for which the number of variables involved is manageable. These methods are often deployed after common parametric specifications are found to be unsuitable for the problem at hand, particularly when formal rejection of a parametric model based on specification tests yields no clues as to the direction in which to search for an improved parametric model.

The job market understood the importance of the non/semi-parametric methods and almost any serious software contains the principal techniques in this area. We illustrate the different models and techniques with R and Matlab. First, R because of the huge number of packages from CRAN, and secondly Matlab because is the easiest environment for programming arrays in econometrics (and typically all objects are arrays in applied econometrics). Both are very representative for the job market. Each lecture will be accompanied by numerical examples and small programming tutorials.

**Course outline :**

- Brief Review of Statistics and Probability Concepts used in the course (pre-requisites)
- Empirical non parametric estimators : histogram, empirical distribution, regressogram
- Kernel-based methods in Non Parametric Econometrics : Parzen-Rosenblatt, Nadaraya-Watson
- Semi-parametric methods : Single index and Additive models
- Splines-based methods in Non Parametric Econometrics : Smoothing splines and Least-Squares splines
- Series-based methods in Non Parametric Econometrics : Wavelets, Laguerre and Hermite Polynomials

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Multivariate and non-linear time series

## CONTENT

Make the students able to cope with multivariate, nonlinear time series analysis.

__Course outline :__

- Introduction to time series analysis
- Nonlinear models (TAR, STAR, MS, tests, forecasting
- Multivariate models (VAR, SVAR, cointegration
- Factor analysis (Macro, fundamental, PCA)
- Multivariate volatility models (VEC, BEKK, DCC, OGARCH)

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Advanced econometrics II: theory and applications (6 ECTS)

### Methodology of econometrics and statistical studies

## CONTENT

Provide students with a set of rules to be followed in the course of realization of a statistical or econometric study for an organization (be it a private company, a public or private administration, a NGO, etc.).

Course outline :

- Starting from the beginning : calls for tender, mutual agreement contracts and within organization projects.
- Identifying the question(s) to be answered. Modelling issues.
- Going from the economic model to the econometric model.
- Collecting, inspecting and managing the data.
- Choosing the right statistical or econometric estimation method(s).
- Analysing the results and their consequences in terms of decision/policy. Professional ethics.
- Communicating about the results in oral and written forms.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Advanced econometrics

## CONTENT

The goal of this course is to present advanced methods in econometrics for distributional analysis, regression and classification models. The course will present theoretical foundations and underlying intuition of each method, as well as several empirical examples.

**Course outline :**

1. Resampling Methods

- Pseudo-random generator
- Monte Carlo experiments
- Bootstrap and permutation tests

2. Nonparametric Econometrics

- Density estimation
- Regression splines
- Finite mixture models

3. Econometrics and Machine Learning

- Philosophy and general principle
- Resampling-based methods and algorithms
- Misspecification detection

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### End-of-study project (6 ECTS)

### Big Data (6 ECTS)

### IT tools for Big Data, a deeper view

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Advanced machine learning

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Applications for Big Data: elective teaching units, choose 2 among 4 (6 ECTS)

### Big data and public policies (3 ECTS)

### Big data and public policies

## CONTENT

The objective is to introduce several questions associated with the use of big data to understand the context, impacts and limits of this new technology. Issues will be presented to students : e.g. access, security, innovation and opportunities to specifically identify how big data can boost local development and allow to design new public policies.

**Course outline :**

Part I : Data and local development

I. Open data, Smart Region and innovation

II. Contracts and legal aspect of the platform FlexGrid, a tool for energy transition

Part II : Big data and security

I. Public security and cities security

II. Cybersecurity

Part II : Big data, a decision-making support

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big data and quantitative marketing (3 ECTS)

### Big data and quantitative marketing

## CONTENT

Understand how data analytics, machine and deep learning methods on large volumes of structured or unstructured data allow to better model, predict or describe consumer behaviour. Understand how data analytics based modelling impacts marketing decisions and how recent advances are changing market research and marketing science.

**Course outline :**

- The 3 main uses of data in marketing analytics :

o Model and describe behaviours / segment consumers o Predict behaviours based on a set of variables o Receive real-time information on consumer behaviours

- The types of data used in quantitative marketing :

o Aggregate statistical/econometric data and time series o Individual observations in data sets (survey data, databases, etc.) o User generated content (text and images) o Datafied objects

- How data is changing and why : from solicited to unsolicited, from structured to unstructured, from manual datafication to algorithm-led datafication, from text to data, etc.
- The main statistical methods used in data analytics
- The code languages used in advanced data analytics
- Algorithms, machine learning, deep learning and artificial intelligence

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big data and finance (3 ECTS)

### Big data and finance

## CONTENT

The course presents the last developments around the use of big data technics in finance. The first part offers an overview of the various recent applications to corporate finance and financial regulation. The second concentrates on the use of big data and associated models in market finance. The third and last part highlights the role of these methods in insurance and reinsurance market.

**Course outline :**

Part 1 Overview of the applications of Big data in finance (6h, Pierre Bittner)

1– The interest of big data in finance

1.1 – Reminder on Big data

1.2 – Big data and decision

1.3 – Big data and market supervision

2– Case study in finance

2.1 – Applications to corporate and investment banking

2.2 – Regulatory challenges

Part 2 : Big data and market finance (12h, Yoann Bourgeois)

1- Realized Volatility

1.1- Continuous time pricing fundamentals

1.1.1 Brownian motion and random walk

1.1.2 Stochastic Differential Equation/ Stochastic Integrals

1.1.3 Quadratic Variation

1.1.4 Implied volatility in Black Scholes

1.2- Realized Volatility

1.2.1 Unbiased estimators

1.2.3 Confidence intervals

1.2.3 Application FX market

1.3-RV and integrated variance

1.3.1 Seasonality

1.3.2 The impact of periodic events on the RV.

1.3.3 Application FX market

2- Bonds portfolio automatic engine

2.1 Definitions (Yields, Bond, Duration, P&L of a bond etc.)

2.2 Bonds clustering (PCA+KMeans)

2.3 The reference curve construction

2.3.1 Regression

2.3.2 Cubic Splines

2.4 Z-Score and momentum to sort bonds

2.5 Reference bonds replication

2.6 Application France 10Y reference bond.

3- Intraday hedging of FX options

3.1 SABR model

3.2 Gatheral parametric local volatility model

3.3 Intraday model calibrations

3.4 Tichonov Regularization

3.5 The use of risk neutral distribution quantiles and moments

3.6 Application FX vanilla options

Part 3 : Big data and insurance (6h, Serdar Coskun)

This part presents, via recent cases, the utility of big data and insurance and reinsurance markets. It also presents the most recent progress in the « insurtech » market.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Big data: other applications (3 ECTS)

### Big data: other applications

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### S4 M2 EBDS magistère option (SE) (36 ECTS)

### Big data IV (6 ECTS)

### Managing Big Data with SAS

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Hands-on project

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Advanced methods in Big Data (9 ECTS)

### Automatic model selection methods

## CONTENT

The objective of this course is to introduce quantitative methods allowing to reduce information. These methods cover different fields of statistics and are based on classical econometric methods (OLS, MLE) or classificatory or principal component methods. The ultimate goal is to study methods to do automatic variable selection in large-scale problems and to apply them to real data.

**Course outline :**

- Classification methods
- Economic factor models
- Statistical factor models
- Lasso methods
- The so-called « General to Specific » method (Hendry, Gets or Autometrics Methodology)

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Predictive methods

## VOLUME OF TEACHINGS

- Lectures:
**24**hours

### Machine learning and statistical learning

## CONTENT

This course provides a broad introduction to statistical learning and machine learning. The main objective is to provide students with the knowledge necessary to understand machine learning methods.

**Course outline :**

Introduction : Statistical learning and machine learning : what and why ?

Part 1. Unsupervised learning

Clustering

Expectation maximization

Dimensionality reduction

Part 2. Supervised learning : regression

Linear regression

Logistic regression

Part 3. Supervised learning : machine learning

Generative models

Bias/variance tradeoff

Cross-validation

Variable selection

Part 4. Machine learning algorithms by hand

Classification and regression trees, Bagging, Random Forest, Boosting, Support Vector Machine, Neural Networks, etc.

## VOLUME OF TEACHINGS

- Lectures:
**24**hours