Home > Research > Publications & Outputs > Extremal financial risk models and portfolio ev...
View graph of relations

Extremal financial risk models and portfolio evaluation

Research output: Contribution to journalJournal articlepeer-review

Published

Standard

Extremal financial risk models and portfolio evaluation. / Zhang, Z; Huang, J.

In: Computational Statistics and Data Analysis, Vol. 51, No. 4, 15.12.2006, p. 2313-2338.

Research output: Contribution to journalJournal articlepeer-review

Harvard

Zhang, Z & Huang, J 2006, 'Extremal financial risk models and portfolio evaluation', Computational Statistics and Data Analysis, vol. 51, no. 4, pp. 2313-2338. https://doi.org/10.1016/j.csda.2006.09.042

APA

Zhang, Z., & Huang, J. (2006). Extremal financial risk models and portfolio evaluation. Computational Statistics and Data Analysis, 51(4), 2313-2338. https://doi.org/10.1016/j.csda.2006.09.042

Vancouver

Zhang Z, Huang J. Extremal financial risk models and portfolio evaluation. Computational Statistics and Data Analysis. 2006 Dec 15;51(4):2313-2338. https://doi.org/10.1016/j.csda.2006.09.042

Author

Zhang, Z ; Huang, J. / Extremal financial risk models and portfolio evaluation. In: Computational Statistics and Data Analysis. 2006 ; Vol. 51, No. 4. pp. 2313-2338.

Bibtex

@article{6119d4f201114d8ea8d16bb04293dbfe,
title = "Extremal financial risk models and portfolio evaluation",
abstract = "It is difficult to find an existing single model which is able to simultaneously model exceedances over thresholds in multivariate financial time series. A new modeling approach, which is a combination of max-stable processes, GARCH processes, and Markov processes, is proposed. Combining Markov processes and max-stable processes defines a new statistical model which has the flexibility of modeling cross-sectional tail dependencies between risk factors and tail dependencies across time. The new model also models asymmetric behaviors of negative and positive returns on financial assets. An important application of the proposed method is to calculate value at risk (VaR) and evaluate portfolio combinations under VaR constraints. Result comparisons between VaRs based on the new approach and VaRs based on some existing methods such as variance–covariance approach and historical simulation approach suggest that some existing methods substantially underestimate the risks during recession and expansion time.",
keywords = "Extreme value theory, M4 processes , Markov chains , Tail dependence index , Financial risk , Portfolio evaluation",
author = "Z Zhang and J Huang",
year = "2006",
month = dec,
day = "15",
doi = "10.1016/j.csda.2006.09.042",
language = "English",
volume = "51",
pages = "2313--2338",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",
number = "4",

}

RIS

TY - JOUR

T1 - Extremal financial risk models and portfolio evaluation

AU - Zhang, Z

AU - Huang, J

PY - 2006/12/15

Y1 - 2006/12/15

N2 - It is difficult to find an existing single model which is able to simultaneously model exceedances over thresholds in multivariate financial time series. A new modeling approach, which is a combination of max-stable processes, GARCH processes, and Markov processes, is proposed. Combining Markov processes and max-stable processes defines a new statistical model which has the flexibility of modeling cross-sectional tail dependencies between risk factors and tail dependencies across time. The new model also models asymmetric behaviors of negative and positive returns on financial assets. An important application of the proposed method is to calculate value at risk (VaR) and evaluate portfolio combinations under VaR constraints. Result comparisons between VaRs based on the new approach and VaRs based on some existing methods such as variance–covariance approach and historical simulation approach suggest that some existing methods substantially underestimate the risks during recession and expansion time.

AB - It is difficult to find an existing single model which is able to simultaneously model exceedances over thresholds in multivariate financial time series. A new modeling approach, which is a combination of max-stable processes, GARCH processes, and Markov processes, is proposed. Combining Markov processes and max-stable processes defines a new statistical model which has the flexibility of modeling cross-sectional tail dependencies between risk factors and tail dependencies across time. The new model also models asymmetric behaviors of negative and positive returns on financial assets. An important application of the proposed method is to calculate value at risk (VaR) and evaluate portfolio combinations under VaR constraints. Result comparisons between VaRs based on the new approach and VaRs based on some existing methods such as variance–covariance approach and historical simulation approach suggest that some existing methods substantially underestimate the risks during recession and expansion time.

KW - Extreme value theory

KW - M4 processes

KW - Markov chains

KW - Tail dependence index

KW - Financial risk

KW - Portfolio evaluation

U2 - 10.1016/j.csda.2006.09.042

DO - 10.1016/j.csda.2006.09.042

M3 - Journal article

VL - 51

SP - 2313

EP - 2338

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

IS - 4

ER -