Back to search
University of Macedonia Economic and Social Sciences
Πολυμεταβλητή ανάλυση χρονικών σειρών
Abstract
dc:descriptionIn this dissertation we examine the problem of classification analysis for time series data from its predictive aspect. We suggest two classification functions for time series through a linear stale-space model representation and, also, for non-linear time series of the GARCH family. Our classification functions are based on the likelihood ratio and the Kullback-Leibler information measure. We propose asymptotic distributions and we investigate their behaviour through simulation experiments.
Degree
thesis:*- Grantor dc:publisher
- University of Macedonia Economic and Social Sciences
- Year dc:date
- 2008
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kalantzis, Thomas
- Καλαντζής, Θωμάς
Subjects
dc:subject × 9Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/30872