| Course language | Frequency | Hours per week | ECTS | Exam |
|---|---|---|---|---|
| English | Winter semester | 4L+2E | 10 | Presentations Intermediate exam (45 minutes) Written exam (90 minutes) |
In empirical economic research, the choice of econometric model and estimation method depends heavily on the research question and the available data. While many econometric methods have now been implemented in econometric software, their application and the interpretation of results generally require a sound knowledge of econometric theory.
The focus is particularly on the properties of the least squares estimator (KQ/OLS estimator) and the generalised least squares estimator (GLS estimator). At the beginning, the basics of econometric models are presented, including which variables can and cannot be modelled using multiple regression. The essential basics of linear algebra for understanding the KQ estimator are covered, as well as the basics of probability theory for analysing the exact and approximate estimation properties of many estimators, in particular the KQ estimator. Furthermore, the properties of common methods for testing (economics) hypotheses are analysed in detail. Finally, the generalised KQ estimator and its application to linear regression models with conditionally heteroscedastic errors are discussed.
! Important !
The free statistical software R (external link, opens in a new window) is used to apply the methods in practice. Students are strongly recommended to attend the elective course Programming with R, which takes place during the lecture period. Important basics are discussed in the lecture, but a deeper insight can be gained more easily in the elective course.
CONTENTS
Methods of econometrics
Preliminary mathematicians course
- Linear algebra (vectors, vector spaces, Euclidean space, matrix algebra)
- Fundamentals of probability theory (random variables, distribution and density functions, moments, conditional probabilities and moments)
- Important probability distributions (standard normal distribution, multivariate normal distribution, chi-square distribution, F-distribution, t-distribution)
- Tools of probability theory for asymptotic analysis: convergence of sequences of random variables (convergence in probability (plim), convergence in distribution, Slutsky's theorem, theorem on continuous mappings)
Main course: Econometric methods
- Basics of estimation and test theory
- Tools for asymptotic analysis (laws of large numbers, central limit theorems)
- Least squares estimators: derivation and geometric interpretation (projections, Frisch-Waugh-Lovell theorem)
- Exact statistical properties of the KQ estimator in finite samples (conditions for unbiased and normally distributed KQ estimators, variance-covariance matrix of parameter estimators, efficiency, Gauss-Markov theorem)
- Approximate statistical properties of the KQ estimator in finite samples using asymptotic methods (exogenous and predetermined regressors, consistency, asymptotic normal distribution partly with derivatives)
- Properties of the KQ estimator for misspecified regression models (bias, mean square error)
- Exact tests(t-test, F-test, Chow structural break test)
- Asymptotic tests(t-test, F-test)
- Empirical applications
- Bootstrap tests and confidence intervals
- Analysing time series data (autocorrelation, stochastic processes, stationarity, autoregressive processes, dynamic linear regression models)
- Heteroscedasticity and autocorrelation robust standard errors (HAC standard errors)
- (Applicable) generalised KQ estimators (GLS, FGLS estimators) and application to heteroscedastic errors
- Tests for model verification
- Outlook on further estimation methods: Instrument variable estimator (IV), maximum likelihood estimator (ML)
LITERATURE
Required literature:
Davidson, R. and MacKinnon, J.G. (2004). Econometric Theory and Methods. Oxford University Press.
Corrections since publication (external link, opens in a new window)
(The book is available several times in the library, is in stock at Pustet on campus or can be ordered there)
Supplementary literature and in-depth literature: see course materials
SOFTWARE
The free software R is used in the course. See note above.
TARGET GROUP / REQUIREMENTS
Compulsory course is part of the compulsory module Methods of Economics in the Master of Economics and IVWL.
Necessary: Knowledge of the first two parts of the preliminary maths course.
For students who have not yet attended an introductory econometrics course during their previous (Bachelor's) degree programme:
Helpful: knowledge of the Bachelor's course Introduction to Econometrics (external link, opens in a new window).
ALLOCATION OF GRADES
The overall grade in the respective course is calculated from the written examination and the coursework completed during the semester. To pass the course, students must pass the written exam and achieve an overall grade of no worse than 4.0. For details see GRIPS.
