Learning outcomes

At the end of the class,  students will be able to distinguish between econometrics methods dealing with different types of dependent variables (continuous, discrete, strictly positive, repeated over individuals).

Secondly, they will be able to estimate and interpret these models using Stata software.

Finally, they will be familiarized with the treatment effect literature and will be able to estimate such models under different sets of assumptions about the treatment effect.

 

Goals

This class aims to familiarize the students with the treatment of various types of datas. As such, different econometrics techniques will be presented, covering cross-sectional and panel data. Besides, we will also cover qualitative dependent variables, which require particular econometric modelling and interpretations.

A second objective is to allow students to be autonomous in the analysis of these data. To achieve this, tutorials, monitored by 2 teaching assistants will be organized. These sessions will focus on applying the methods seen in class on real data using the Stata software. Besides, they can serve as discussion room for the homework that has to been handed at the end of the period.
 
The topics that are covered are a combination of subjects typically covered in introductory and more advanced courses in econometrics. While some background in econometrics is helpful, the course is also open to students without prior knowledge of econometrics (but willing to study).



 

Content

This class will cover different topics, giving the students the opportunity to be in touch with many different econometric specifications, depending on the data they have to handle.The material not covered during the lectures is not part of the final exam.

1) Refresher on linear models
 
2) Endogeneity
 
3) Causal inference models

2) GMM and Maximum Likelihood estimation frameworks

5) Poisson regression
 
6) Panel data

7) Qualitative dependent variables
 

 

Exercices

6 tutorials will be organized to help students to apply the theoretical concepts seen in class on real data using the Stata software.

These sessions will be monitor by 2 different teaching assistants.
 

Assessment method

For the first session exam, the evaluation consists of two parts. 1) A homework in which the student needs to define a research question, find an appropriate database and apply the appropriate techniques to answer the research question (30%).

2) A written exam (70%).

For the september session, the final grade will only consist of the grade of the final exam

 
Rules concerning the use of Artificial Intelligence tools
This course sticks to three out of the four key principles set out in the UClouvain note on the use of AI tools:
 
 - Principle of responsibility: Students are entirely responsible for the work they submit. They must be the result of a personal process.
- Principle of transparency: Where necessary, students should clearly indicate the aids and tools used. Such use must respect the principles of academic integrity.
- Principle of authenticity: Students must ensure that their work enables an assessment of the knowledge and skills they have authentically acquired. This applies both to the result of their work and to the process and method used to produce it.

For this course, you're only allowed to use AI tools to translate your homework to english and/or improve the english writing. If you do so, you need to explicitly mention the use of AI tools (see the UCLouvain note for more details).

Sources, references and any support material

The course borrows material from several textbooks. The main one is

Wooldridge, J., 2010, Econometric Analysis of Cross section and Panel Data, The MIT Press, London

The support material will consists of the slides posted on Webcampus.

 

Language of instruction

Français