Universität zu Köln

Pair Constructions for High-Dimensional Dependence Models in Discrete and Continuous Time

Nicklas, Stephan (2013) Pair Constructions for High-Dimensional Dependence Models in Discrete and Continuous Time. PhD thesis, Universität zu Köln.

[img]
Preview
PDF
Download (1895Kb) | Preview

    Abstract

    Modeling high-dimensional dependence structures with parametric dependence models is a challenging and important part in statistics. The dependence models that we discuss here are based on bivariate building blocks and we illustrate how to assemble these building blocks to multivariate dependence structures. In the first part of the dissertation, we recall the pair-copula construction for random variables, present different extensions and apply the concept in an empirical study. In the second part, we introduce the new pair-Lévy copula construction to model the jump dependence of high-dimensional Lévy processes. Moreover, we present simulation and estimation algorithms, conduct a simulation study and discuss various applications of this concept.

    Item Type: Thesis (PhD thesis)
    Creators:
    CreatorsEmail
    Nicklas, Stephannicklas@wiso.uni-koeln.de
    URN: urn:nbn:de:hbz:38-51324
    Subjects: General statistics
    Uncontrolled Keywords:
    KeywordsLanguage
    Pair-copula construction, Vine copula, Lévy copula, Pair-Lévy copula construction, Multivariate Lévy processesEnglish
    Faculty: Wirtschafts- u. Sozialwissenschaftliche Fakultät
    Divisions: Wirtschafts- u. Sozialwissenschaftliche Fakultät > Seminar für Wirtschafts- und Sozialstatistik
    Language: English
    Date: 15 May 2013
    Date Type: Publication
    Date of oral exam: 10 May 2013
    Full Text Status: Public
    Date Deposited: 11 Jun 2013 15:02:38
    Referee
    NameAcademic Title
    Schmid, FriedrichUniv.-Prof. Dr.
    URI: http://kups.ub.uni-koeln.de/id/eprint/5132

    Actions (login required)

    View Item