Javascript must be enabled for the correct page display

Modelling the dependence structure of financial time series: considering copula-MGARCH models

Chu, H. (2017) Modelling the dependence structure of financial time series: considering copula-MGARCH models. Master's Thesis / Essay, Mathematics.

[img]
Preview
Text
Hung_Chu_2017_WM.pdf - Published Version

Download (1MB) | Preview
[img] Text
Toestemming.pdf - Other
Restricted to Backend only

Download (80kB)

Abstract

In this thesis we study multivariate models using twenty financial time series. The models are obtained by combining copula functions with multivariate GARCH models. The advantage is that these models can account for the volatile behaviour of the time series through the multivariate GARCH specification, while allowing the copula functions to explicitly account for the underlying dependence structure. The dependence structure of the time series will be described by a rank correlation and tail-dependence coefficients. The main objective of this thesis is to identify adequate models for fitting and forecasting of multivariate time series data and how these models describe the underlying dependence structure. We achieve this by evaluating the models using three different tests: one for model selection, one for model validation and one for model ranking. For comparison, we also evaluate traditional multivariate models. By combining the findings of the three tests, we conclude that the DCC-ARMA-GJR model incorporated with the $t$ copula is the most appropriate.

Item Type: Thesis (Master's Thesis / Essay)
Degree programme: Mathematics
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 15 Feb 2018 08:27
Last Modified: 15 Feb 2018 08:27
URI: https://fse.studenttheses.ub.rug.nl/id/eprint/15126

Actions (login required)

View Item View Item