Dr Marcia Schafgans, 32L 4.12

Professor Steve Pischke, 32L.2.16

This course is compulsory on the BSc in Econometrics and Mathematical Economics, BSc in Economics and MSc in Economics (2 Year Programme). This course is not available as an outside option nor to General Course students.

Students must have completed Elementary Statistical Theory I (ST109), Quantitative Methods (Mathematics) (MA107) or Mathematical Methods (MA100), and Econometrics I (EC1C1), or equivalent.

MSc in Economics (2 Year Programme) students can take Mathematical Methods (MA100) alongside EC2C1.

This course builds on the knowledge learned in Econometrics I (EC1C1). In the MT part of the course, the focus is an introduction to the theory of econometrics. You will study in detail various estimators common in the literature: the least squares estimator, the instrumental variable estimator, and the maximum likelihood estimator. You will discuss how (and whether) these estimators can be used for inference purposes under a range of assumptions underlying the data generating process. Topics include the derivation of finite sample properties (unbiasedness, precision (standard error) and efficiency), asymptotic properties (consistency and asymptotic distribution), confidence intervals and hypothesis testing. Most discussions will pertain to the use of cross-sectional data and the linear model. We will consider the binary choice model as an example of a nonlinear model and will cover some aspects of time series data. Some discussions make use of matrix algebra.

In the LT part of the course, you will learn more econometric techniques, including difference-in-differences, and regression discontinuity designs that make use of panel data and you will revisit the instrumental variable technique. Teaching in LT will be based on empirical examples and you will find out how to conduct your own empirical investigation. As part of the course, you will work on an empirical project and write an individual report about your analysis and findings.

30 hours of lectures, 10 hours of classes and 10 hours of help sessions in the MT. 20 hours of lectures and 9 hours of classes in the LT. 1 hour of classes in the ST.

There will be a reading week in Week 6 of LT, during which there will be no lectures or classes.

Student learning will be supported through the EC2C1 Support Lab and through a dedicated discussion forum.

During MT, there are weekly assignments and feedback will be given on two.

During LT, there will be three homework assignments. Students are expected to give a progress report on their individual project and are given feedback on this.

• J. Wooldridge Introductory Econometrics. A Modern Approach, Cengage

• J. D. Angrist and J. S. Pischke Mastering ‘Metrics. The Path from Cause to Effect, Princeton University Press.

Exam (50%, duration: 2 hours, reading time: 15 minutes) in the January exam period.
Coursework (50%, 2500 words) in the ST.