Managing portfolio risk using multivariate extreme value methods

Arbeitspapier

This paper provides a strategy for portfolio risk management by inferring extreme movements in financial markets. The core of the provided strategy is a statistical model for the joint tail distribution that attempts to capture accurately the data generating process through an extremal modelling for the univariate margins and the multivariate dependence structure. It takes into account the asymmetric behavior of extreme negative and positive returns, the heterogeneous temporal and cross-sectional lead-lag extremal dependencies among the portfolio constituents. The strategy facilitates scenario generation for future returns, estimation of portfolio profit-and-loss distribution and calculation of risk measures, and hence, enabling us to answer several questions of economic interest. We illustrate the usefulness of our proposal by an application to stock market returns for the G5 economies.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: Manchester Business School Working Paper ; No. 636

Classification and Topics

• Classification: Wirtschaft ; Semiparametric and Nonparametric Methods: General ; Statistical Simulation Methods: General ; Multiple or Simultaneous Equation Models: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes; State Space Models
• Subject: ARMA-GARCH filtering ; Asymptotic dependence ; Asymptotic independence ; Copula ; Multivariate extreme values

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Hilal, Sawson ; Poon, Ser-Huang ; Tawn, Jonathan
• Event: Veröffentlichung
• (who): The University of Manchester, Manchester Business School
• (where): Manchester
• (when): 2013

Further information

• Handle: http://hdl.handle.net/10419/102368
• Last update: 01.04.2025, 12:44 PM CEST