Παρουσίαση/Προβολή

(STAT216) -  Ioannis Vrontos

Περιγραφή Μαθήματος

This course provides the theory and practice of time series analysis. The first part introduces deterministic models. The key concept in traditional time series is the decomposition of a given time series into its components: a trend, a seasonal and an irregular component. The course presents and describes different models for estimating trend and seasonal effects. It introduces linear filters and exponential smoothing techniques. The second part introduces stochastic time series models. After introducing the basic theory of stationary processes, it describes and presents analytically the Box-Jenkins methodology for ARIMA models. The course introduces the class of conditional heteroscedastic models (ARCH/GARCH), and presents practical time series forecasting techniques. Illustrative examples applying time series models/techniques to actual economic and financial data are also presented using the econometric package Eviews, and also using R package. The empirical analysis consists of (a) unit root testing to exchange rate series and financial series, e.g. stocks and indices, and (b) modeling and forecasting economic/financial return series.

Ημερομηνία δημιουργίας

Πέμπτη, 9 Φεβρουαρίου 2017