Zur Merkliste hinzufügen


30661-01 - Vorlesung: Advanced Time Series Analysis 3 KP

Semester Frühjahrsemester 2021
Angebotsmuster Jedes Frühjahrsem.
Dozierende Simon Beyeler (simon.beyeler@unibas.ch)
Sylvia Kaufmann (sylvia.kaufmann@unibas.ch, BeurteilerIn)
Inhalt The course introduces multivariate time series modelling, in particular vector autoregression. In a first part, we discuss the model under an abstract viewpoint. We derive properties of the time series process, discuss stationarity and invertibility conditions, derive conditional and unconditional moments, identify sources of uncertainty. As single parameters are not of prime interest, tools like impulse responses and variance decomposition are used to interpret multivariate time series models. Besides improved structural in-sample analysis, multivariate analysis may improve eventually forecasting performance.

Two following parts discuss model estimation from a frequentist and a Bayesian perspective, respectively. A Bayesian approach circumvents estimation difficulties when either data information is scarce (i.e. data series are short) or large (many time series are available). We discuss model choice and specification measures, and procedures to quantify uncertainty.

The last part introduces latent variables into multivariate modelling. These capture latent processes determining the data like regime-switching parameters or underlying common factors.

The course includes some exercise sessions.
Lernziele - Analyse and derive the properties of multivariate time series models.
- Perform model specification/comparison; understand and apply tools to interpret model estimates.
- Understand the differences between frequentist estimation and Bayesian inference.
- Implement estimation and various structural identification procedures, quantify uncertainty.
- Estimate basic latent variable models.
- Basic knowledge of forecasting procedures, forecast evaluation.
Literatur Gelman A., Carlin J.B., Stern H.S. and Rubin, D.R. (1995), Bayesian Data Analysis, Chapman and Hall, London.
Lütkepohl Helmut, 2005, New Introduction to Multiple Time Series Analysis, Springer.
Neusser Klaus, 2016, Time Series Econometrics, Springer International Publishing AG Switzerland.
Weblink Weblink


Teilnahmebedingungen Completed BA (preferably in economics).
Econometrics MA level; univariate time series modelling (advantageous).

Knowledge of econometrics or programming software (like e.g. EViews, matlab, R)
Anmeldung zur Lehrveranstaltung Registration: Please enrol in MOnA. EUCOR-Students and students of other Swiss Universities have to enrol at the students administration office (studseksupport1@unibas.ch) within the official enrolment period. Enrolment = Registration for the exam!
Unterrichtssprache Englisch
Einsatz digitaler Medien kein spezifischer Einsatz


Intervall Wochentag Zeit Raum
14-täglich Mittwoch 14.15-18.00 Wirtschaftswissenschaftliche Fakultät, Auditorium
Bemerkungen The course will be taught online at the dates you can see below and then continue in-class with a simoultaneous live-stream as soon as it is possible.


Datum Zeit Raum
Mittwoch 24.03.2021 14.15-18.00 Uhr - Online Präsenz -, --
Mittwoch 31.03.2021 14.15-18.00 Uhr - Online Präsenz -, --
Mittwoch 07.04.2021 14.15-18.00 Uhr - Online Präsenz -, --
Mittwoch 21.04.2021 14.15-18.00 Uhr - Online Präsenz -, --
Mittwoch 05.05.2021 14.15-18.00 Uhr - Online Präsenz -, --
Mittwoch 19.05.2021 14.15-18.00 Uhr Wirtschaftswissenschaftliche Fakultät, Auditorium
Mittwoch 02.06.2021 14.15-18.00 Uhr Wirtschaftswissenschaftliche Fakultät, Auditorium
Module Modul: Fachlich-methodische Weiterbildung (Doktoratsstudium - Wirtschaftswissenschaftliche Fakultät)
Modul: Statistik und Computational Science (Masterstudium: Actuarial Science)
Spezialisierungsmodul: Areas of Specialization in International and/or Monetary Economics (Masterstudium: International and Monetary Economics)
Vertiefungsmodul: Quantitative Methods (Masterstudium: Wirtschaftswissenschaften)
Leistungsüberprüfung Semesterendprüfung
Hinweise zur Leistungsüberprüfung written exam: 29.06.21; 10:15-11:45. The exam will take place at WWZ. In case COVID-19 protective measures prevent examination on site, the faculty reserves the right to conduct the examination electronically during the same time slot. You will receive details of the on-site examinations (Exhibition Center or WWZ) by email approximately one week before the examination date.
An-/Abmeldung zur Leistungsüberprüfung Anmeldung: Belegen
Wiederholungsprüfung keine Wiederholungsprüfung
Skala 1-6 0,1
Wiederholtes Belegen beliebig wiederholbar
Zuständige Fakultät Wirtschaftswissenschaftliche Fakultät / WWZ, studiendekanat-wwz@unibas.ch
Anbietende Organisationseinheit Wirtschaftswissenschaftliche Fakultät / WWZ