Course content
High dimensional data sets are increasingly available for the analysis of economic, business, and finance problems. As these data are normally generated by usual business activity or operation, they originate from real business or economic processes that are not compatible with the standard assumptions of statistical regression models. By breaching these assumptions, alternative econometric models need to be employed that can provide valid estimation and inference results in these scenarios and are tailored to uncover the partial or causal relationship between economic variables.
The course gives students an understanding of elementary econometric regression models which are often used in economics and finance to analyse data sets. The course introduces the material from both a theoretical and practical angle. It contains formal treatment of statistical assumptions and properties of estimators using matrix notation. It also presents applications with real data from economics and finance, where students learn how to use the statistical software R to obtain interpretable results. Strong emphasis is put on explaining the link between the statistical theory and empirical practice. Students eventually learn how the models and their restrictions translate into practical work with the statistical software R.
The course consists of lectures and exercise classes. The lectures are followed by computer classes, where students deepen their understanding by working on theoretical and empirical problems. Students can work in groups to solve the weekly problem sets and present their solutions to obtain feedback.
The course has two parts:
A Multiple regression analysis
B Endogeneity and non-linear models
A Multiple regression analysis
A0 Intro: What is econometrics? (1h)
A1 Estimation by OLS, Properties, Gauss-Markov, variable choice (5h)
A2 Violations of Gauss Markov- assumptions (heteroskedasticity, serial correlation, (F)GLS) (4h)
A3 Policy analysis (2h)
MIDTERM EXAM
B Topics in cross section econometrics
B1 Endogeneity (4h)
B2 Simultaneous equation models (2h)
B3 Maximum likelihood estimation (2h)
B4 Limited dependent variable models (3h)
B5 Evaluation and review (1h)
FINAL EXAM
See course description in course catalogue