Time Series Analysis

Guest Instructor: 
Amanda Fernández Fontelo
Time I: 
Wednesday,
12:00pm to 02:00pm
Time II: 
Thursday,
08:00am to 10:00am
Venue I: 
HU Berlin, Spandauer Str. 1, Room 21a
Venue II: 
HU Berlin, Spandauer Str. 1, Room 25
Description: 

Information on how to attend the online course will be available on Moodle.

The course aims at providing the basic concepts and methods for analysing time series data. The focus is on univariate modelling tools. The lecture begins with classical components models. Then we cover different types of stochastic processes like ARIMA and GARCH models, deal with the unit root methodology and procedures for forecasting as well as for the specification, estimation and validation of models. Multivariate extensions are demonstrated, with emphasis on vector autoregressive (VAR) processes and its application in causality and impulse response analyses. Nonstationary systems with integrated and cointegrated variables will also be treated. In the last session, a brief introduction to count time series, with particular emphasis in INAR(1) models and their applications, will be introduced.

In the tutorials the time series methods are applied to empirical data. We will intensively make use of econometric software packages.

Classical components models; stochastic processes; stationarity; ARIMA processes, GARCH models; specification, estimation and validation of models; forecasting; unit root tests; multivariate extensions: VAR processes, causality and impulse response analysis, cointegrated processes. In the tutorials the time series methods are applied to empirical data.

Literature:
Hamilton, D.J. (1994). Time Series Analysis, Princeton University Press.
Lütkepohl, H. (2005). New Introduction to Multiple Time Series Analysis, Springer Verlag, Heidelberg

Exam:
written exam (90 min)

Credits: 
6.00
Program: 
Semester: 
Spring 2020
Affiliation: 
Humboldt-Universität zu Berlin