| Course language | Frequency | Hours per week | ECTS | Exam |
|---|---|---|---|---|
| German | Summer semester | 2L+2E | 6 | Intermediate exam (30 minutes) Final exam (90 minutes) |
Contents
The analysis of time series data plays a central role in empirical economic research and, more generally, in data science, and is therefore the focus of time series econometrics. Time series data can exhibit different dynamics and be characterised by trends, seasonal patterns, structural breaks or long-term equilibrium relationships.
Students learn to recognise the respective characteristics of time series data and to select and apply appropriate econometric time series models. They learn the relevant fundamentals of statistical and econometric theory and their practical application based on the freely available software R.
The course Time Series Econometrics deals with the analysis of time series data. This includes studying the properties of possible data-generating time series processes. Of central importance here is the theory of autoregressive processes (think of these as a stochastic variant of deterministic difference equations). Analysing time series data requires a range of additional knowledge, as, for example, the estimation properties of the ordinary least squares (OLS) estimator depend on the exogeneity properties of the regressors or on the dynamic stability properties of the autoregressive process that may have generated the observed data. It is also important whether there is a trend or a seasonal pattern. Closely related to this is the question of whether an observed time trend is deterministic in nature or a realisation of a random walk. So-called unit root tests are necessary to answer this question. If time trends characterise several time series together, there may be long-term equilibrium relationships between the variables, for the modelling of which cointegration models are used. These models play a particularly important role in empirical macroeconomics and are an indispensable tool for forecasting.
Structure
- Time series models with strictly exogenous regressors
- Trends and seasonality
- Autoregressive time series models
- Asymptotic properties of the OLS estimator for autoregressive models
- Non-stationary time series processes, random walk processes
- Dynamic regression models with uncorrelated errors
- Regression models with autocorrelated and heteroscedastic errors
- Tests for autocorrelation in the residuals
- Unit root tests: tests to verify the random walk hypothesis
- Error correction models, cointegration (estimation and test)
- Forecasting and forecasting intervals
Literature
Wooldridge, J.M. (2009 - or newer). Introductory Econometrics. A Modern Approach, 4th edition, Thomson South-Western (Chapters 5, 7, 9, 10 - 12, 18).