• Huber Martin

• Oberhänsli Sarina Joy
• Stoller Andreas

This course discusses several of the practically most relevant econometric/statistical methods for empirical research (e.g. in economics and social sciences) along with their underlying assumptions and properties. It also presents applications of these methods in real-world data using the statistical software “R (studio)”. The course consists of a lecture and 4 PC lab sessions. The topics covered in the lecture include:

1. The difference between causation (e.g. education has a causal effect on wage) and correlation (subjects with higher education have higher wages, but this may be driven by other factors than education as for instance ability); the intuition of experiments for assessing causation.
2. Linear regression (OLS - ordinary least squares) to assess the association of one or several variables (e.g. education, age,...) with an outcome of interest (e.g. wage).
3. Quantile regression to conduct empirical analyses at particular outcome ranks (e.g. for the median earner in the wage distribution).
4. Nonlinear regression (probit regression for binary outcomes like working vs. not working, tobit regression for censored outcomes).
5. Instrumental variable regression and regression discontinuity designs under endogeneity.
6. Panel data regression and “Differences-in-Differences” estimation when subjects are observed at several points in time.
7. Synthetic Control Method.
8. General concepts of estimation: M-estimators, maximum likelihood, and generalized methods of moments.
9. Bootstrapping (resampling from the original data, e.g. for variance estimation).
10. Time series regression, e.g. for modeling stock prices or GDP growth over time.

As practical illustration, the PC lab sessions consider empirical examples, in which the different econometric methods are applied to various data sets using the statistical software R (studio).

The objectives are that students (1) get familiar with the econometric methods most commonly used in applied work, (2) understand the differences in the properties and assumptions of the various methods along with their advantages and disadvantages, and (3) can practically implement econometric methods to analyze real-world data.

Wooldridge, J. M. (2010): Econometric Analysis of Cross Section and Panel Data, 2nd edition, The MIT Press.

Huber, M. (2023). Causal Analysis: Impact Evaluation and Causal Machine Learning with Applications in R. The MIT Press.

Examination time: 90 Min. and participation in PC labs

Examination time: 90 Min. and participation in PC labs