Objectifs

In this course, students will learn how to develop, estimate, interpret and present results from microeconometric models for so-called limited dependent variables.

Limited dependent variable models are econometric models for outcomes that do not have a simple continuous distribution but whose distribution is "restricted in some important way". Limited dependent variables occur frequently in microdata analyses with models for binary outcomes, ordered outcomes, multinomial outcomes, censored outcomes, truncated outcomes and count data outcomes being some examples.

Econometric methods for analyzing limited dependent variable models will be discussed based on conceptual and theoretical considerations as well as via illustrative examples based on actual data from the US Health and Retirement Study (https://hrs.isr.umich.edu/about). Students will learn how to apply these methods based on practical exercises, using R.

Students will also practice how to develop, execute and present the results from an empirical project on a topic of their choice based on the Swiss Household Panel dataset.

Contenus

The course will cover the following topics:

1. Introduction and general approaches to analyzing micro data
2. Maximum likelihood estimation
3. Binary outcomes
4. Multiple outcomes
5. Censored and truncated outcomes
6. Count data models

The course consists of lectures and laboratory sessions. Empirical illustrations and examples discussed in class will be based on the software R.

Références

The main references for the course are the lecture slides. Additional useful material on the topics discussed in class can be found in the following textbooks:

Cameron, A.C. and Trivedi, P.K., 2005. Microeconometrics: methods and applications. Cambridge university press.

Greene, W. H., 2012. Econometric Analysis. Prentice Hall.

Wooldridge, J.M., 2010. Econometric analysis of cross section and panel data. MIT press.

Pré-requis

Solid background in basic econometrics. Prior knowledge of the software R is not needed.

Evaluation

1ère tentative

Examen:

Sans examen (cf. modalités)  

Evaluation:

The final grade is a weighted average of two grades: one for the project (70%) and the second for a multiple-choice questionnaire (30%).

The questionnaire will be completed in class in mid-May (60 minutes).

The project grade will be based on the quality of the presentation slides, the oral presentation, and the codes that the students will hand in.

Rattrapage

Examen:

Sans examen (cf. modalités)  

Evaluation:

The final grade is a weighted average of two grades: one for a revised version of the project (70%) and the second for a multiple-choice questionnaire (30%).