Due to the COVID-19 crisis, the information below is subject to change, in particular that concerning the teaching mode (presential, distance or in a comodal or hybrid format).
The course proposes an up-to-date presentation of methods to model, analyse and test continuous, discrete and partially observed (censored or truncated) cross section and panel data models. For panel data models, both static and dynamic approaches are considered. The course includes an introduction to the treatment of sample selection and attrition. If time permits, some variable open topics such as multinomial and ordered response models, semi-parametric approaches to static models and non-stationary panels can be included in the course according. Econometric modelling and statistical inference are presented at an advanced level. The course also includes an introduction to an econometric software (such as Stata®), which is used to illustrate all methods through applications.
At the end of this learning unit, the student is able to :
|1||This course focuses on the analysis of cross section and panel data. Cross section data refer to a large number of sample units (individuals, firms, ...) at a given instant of time. Panel data refer to the repeated observation of sample units over time. In this course we limit the analysis to panels with a large number of sample units that are followed over a limited number of time periods. The objective of the course is to give the student a firm understanding of the workings of certain methods and a background for developing new methods. In addition, students should learn to implement certain methods by running econometric software programmes on data sets coming from published work. As such, the course in-tends to provide the necessary background both for the applied researcher who wants to uncover some causal relationship between two variables and for the theoretical researcher aiming at the development of new estima-tion methods.|
The contribution of this Teaching Unit to the development and command of the skills and learning outcomes of the programme(s) can be accessed at the end of this sheet, in the section entitled “Programmes/courses offering this Teaching Unit”.
Contents: 1) Introduction. The nature of panel data. 2) Linear, static model for panel data. Estimating System of Equations by OLS and GLS. Modelling individual heterogeneity. Fixed effect, Random effect models. Specification test. First difference meth-ods. Random effect with correlated unobserved effect. The approach of Mundlak and Chamberlain. The GMM approach (Haussman and Taylor types of estimator) 3) Linear, dynamic model for panel data. The bias of usual estimators. Estimator of Anderson-Hsiao, estimator of Arellano-Bond, estimator of Blundell-Bond. 4) Discrete panel data. Refreshing on multiple response model. Random effect models for a binary de-pendent variable.Fixed effect models. Incidental parameter problem. 5) Limited dependent variable and selection models. Truncated/censored data. Tobit I, II and III models. Attrition and selection bias. Random effect Tobit model. Fixed effect in truncated regression
Due to the COVID-19 crisis, the information in this section is particularly likely to change.The evaluation for this course is twofold:
a) a take-home exam that students will have to complete during a period of 2 to 3 weeks. It will consist in 1) reproducing and extending the results of an empirical paper using real data and 2) writing a 2 pages referee report of a research paper. Students will have to submit their take home on Moodle.
b) An oral examination (either at the office of the teacher or on Microsoft Teams) to discuss the solutions of the take home exam.
Advanced econometrics I A take home exam consisting of an empirical implementation of methods covered during the lec-tures, and an oral exam. No specific support. Some mandatory readings from the literature in microeconometrics.
Faculty or entity