Ec524

Academic Year: 2008/2009

Ec524: EMPIRICAL METHODS IN APPLIED ECONOMICS

Module: Nonstructural Identification

Lecturer: Steve Pischke
Room: R.448
Ext: 6509
Email: s.pischke at lse.ac.uk

General Information
This is a 10 hour module primarily intended for 2nd year MRES students, who are interested in research in an applied microeconomics field (e.g. Development Economics, Political Economy, Labour Economics, International Economics) but might be of interest to others as well (e.g. students in Macro, Finance). The course module will cover empirical methods that are commonly used in these fields. It will help students to better understand readings which will be assigned in 500 level courses in these fields, as well as with their own research. The topics in the course will be organized around specific methods, but these methods will be illustrated with concrete examples from the literature. The overriding theme is the estimation of causal or treatment effects. It will be assumed that participating students have attended EC443 Advanced Econometrics or EC402 Methods of Economic Investigation previously. The module is voluntary and non-examinable, but it is expected that students attend the entire module and do weekly assigned readings.

Time and Room
This module will be taught during weeks 2 to 5 of the Michaelmas Term.
Note: There will no lecture during week 1
Thursdays 11:00-13:00
Room: G1

This course page is no longer actively maintained

Please refer to the course page on LSE Moodle
Link to Moodle

Syllabus and Readings

New book:
Joshua Angrist and Jorn-Steffen Pischke Mostly Harmless Econometrics: An Empiricist's Companion, Princeton University Press, 2008

Topics:

1. The Evaluation Problem: Introduction and Randomised Experiments
2. Observational Studies and Regression
3. Instrumental Variables
4. Differences-in-differences, Panel Data, and Fixed Effects
5. Regression Discontinuity Design
6. Standard Error Issues
7. Nonlinearity and Heterogeneity

You can find some lecture notes on measurement error here. These will not be discussed in the lectures but may be of interest to participants.