FAM-ily  

Financial and Actuarial Mathematics  
TU Wien, Austria  

 
2005-03-15

Einladung zur Vortragsreihe aus Finanz- und Versicherungsmathematik

Dr. Alexandra Dias

Credit Suisse and ETH Zurich

Copula change-point detection and dynamic copula models for multivariate high-frequency data in finance

The stylized facts of univariate high-frequency data in finance are well known. In this work, we concentrate on multivariate high-frequency data, and analyse the dependence structure, through the notion of copula. More in particular, we analyse the conditional copula for two dimensional high-frequency data. We find evidence of non-constancy over time of the conditional copula and investigate this fact at six different time aggregation frequencies. Further, we investigate the existence of change-points in the conditional copula. The data analysed are high-frequency foreign exchange spot rates for US Dollar quoted against German Mark and quoted against Japanese Yen.


Alexandra Dias is Credit Risk Analyst with the Department of Credit Risk Management of Credit Suisse since January this year. Before she was RiskLab Research Fellow at the Department of Mathematics of the Swiss Federal Institute of Technology (ETH) in Zurich where she got her Ph.D. in the beginning of 2004. The Ph.D. work was in copula inference for finance and insurance. She has an M.Sc. on Actuarial Science and Financial Risk Management. Multivariate problems within credit risk, risk management, insurance and the corresponding interplay between finance and insurance are her main areas of interest.

Termin: Dienstag, 15. März 2005, 16:30 Uhr s.t.
Ort: Technische Universität Wien
1040 Wien, Wiedner Hauptstraße 8-10
Freihaus, Turm B (grüner Bereich), 2. Stock,
Hörsaal FH 6

Direktor Helmut Holzer
Präsident der Aktuarvereinigung Österreichs
Präsident des Österreichischen Förderungsvereins der Versicherungsmathematik

Gen.-Dir. Dr. Siegfried Sellitsch
Präsident der Österreichische Gesellschaft für Versicherungsfachwissen

o.Univ.-Prof. Dr. W. Schachermayer
Univ.-Prof. Dr. U. Schmock
Finanz- und Versicherungsmathematik (FAM), Technische Universität Wien