CXE_ECSR_M Time series econometrics in R

Faculty of Economics and Administration
Spring 2024
Extent and Intensity
0/0/0. 4 credit(s). Type of Completion: k (colloquium).
Teacher(s)
doc. Ing. Daniel Němec, Ph.D. (lecturer)
Ing. Mgr. Vlastimil Reichel, Ph.D. (assistant)
Mgr. Jakub Chalmovianský, Ph.D. (assistant)
Guaranteed by
doc. Ing. Daniel Němec, Ph.D.
Department of Economics – Faculty of Economics and Administration
Contact Person: doc. Ing. Daniel Němec, Ph.D.
Supplier department: Department of Economics – Faculty of Economics and Administration
Course Enrolment Limitations
The course is offered to students of any study field.
Course objectives
This course is an introductory course in applied time-series econometrics focusing on the principles of dealing with univariate time series by applying ARMA models and the Box-Jenkinson methodology. This course also introduces tools and methods for short-term forecasts and evaluating the quality of prediction models, techniques for decomposing trends, and basic approaches to testing and solving problems associated with structural breaks, seasonality, and unit-root testing. The course uses practical applications of working with time series using the R software (no prior knowledge of R is required).
The course is designed for those who want to learn how to use standard techniques and methods of econometric time-series analysis and use them to perform appropriate empirical real data analysis aimed at short-term predictions of any variables, including related diagnostics of the results obtained. In addition to explaining standard techniques, the course will introduce the actual trends in the time-series econometrics applicable in practice. The course does not require prior knowledge of econometrics, statistics or programming. Graduates of the course will gain a good foundation for carrying out a flawless time-series analysis dealing with estimating and predicting trends, modelling seasonality and structural breaks. Mastering these universally applicable methods represents a tremendous competitive advantage in the labour market. Obtained skills and knowledge can help better understand other modern techniques (such as machine learning methods focused on high-frequency data, etc.). Graduates of the course will also learn more general skills for working with robust and freely available R software.
Learning outcomes
After completion of the course, students should be able to:
- analyse real data practically using a computer;
- formulate an appropriate model for the data;
- construct predictions;
- be able to evaluate and interpret the results obtained;
- be able to understand papers and studies in the field of time-series econometrics.
Syllabus
  • 1. Modelling of univariate time series (basic motivation and use of ARMA models, stability and stationarity of models, autocorrelation and partial autocorrelation functions, Box-Jenkinson methodology)
  • 2. Prediction (one-step ahead prediction, multi-step ahead prediction, prediction errors, evaluation of prediction quality, other prediction error statistics)
  • 3. Seasonality and structural breaks (seasonality and its modelling, structural breaks and their testing)
  • 4. Non-stationary time series and trend modelling (time series trends, basic unit root tests, univariate methods for trend decomposition)
Literature
    required literature
  • HEISS, Florian. Using R for introductory econometrics. 2nd edition. Düsseldorf: Florian Heiss, 2020, 368 stran. ISBN 9788648424364. info
    recommended literature
  • WOOLDRIDGE, Jeffrey M. Introductory econometrics : a modern approach. Seventh edition. Boston: Cengage Learning, 2020, xxi, 826. ISBN 9781337558860. info
  • BÉKÉS, Gábor and Gábor KÉZDI. Data analysis for business, economics, and policy. First published. Cambridge: Cambridge University Press, 2021, xxiii, 714. ISBN 9781108483018. info
  • BROOKS, Chris. Introductory econometrics for finance. Fourth edition. Cambridge: Cambridge University Press, 2019, xxxi, 696. ISBN 9781108422536. info
  • KOOP, Gary. Introduction to econometrics. Chichester: John Wiley & Sons, 2008, 371 s. ISBN 9780470032701. info
  • ENDERS, Walter. Applied econometric time series. 4th ed. Hoboken: Wiley, 2015, x, 485. ISBN 9781118808566. info
Teaching methods
The course is divided into four thematic study blocks (with a time interval of one week, which is intended for self-study), which are then further divided into particular topics (all this is incorporated into the interactive syllabus within the four main chapters and the individual subchapters incorporated into them). A brief introduction to working with R introduces these four basic blocks.
Each subtopic contains illustrative, annotated examples, essential motivation, and intuitive explanations of the issues. The conclusion of each block includes assignments of examples to work on separately.
After each topic block is completed, participants have the opportunity to participate in an online consultation on all the issues and problem sets discussed in the block, which includes an outline of the following topic block (for the first topic block, the online consultation at the beginning of the course serves this purpose).
Assessment methods
Individual completion of assigned tasks (with comments) for each thematic block. To obtain micro-credentials, it is necessary to further prepare and present three examples (assignments) based on the discussed thematic blocks during the online consultation at the end of the course. The obtained microcredentials can be accepted for recognising BKE_CARA Time Series course (for occupational studies).
Language of instruction
Czech
Further Comments
The course is taught only once.
The course is taught: in blocks.
  • Enrolment Statistics (recent)
  • Permalink: https://is.muni.cz/course/econ/spring2024/CXE_ECSR_M