Managing risk with simulated copula

M. Malavasi, R. Previtali, S. Ortobelli, C. Nardelli

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

Abstract

The aim of this study is verify whether the Average Value at Risk (AVaR) can be a good alternative to Value at Risk (VaR), for estimating great portfolio losses, especially regarding tail events. To do so we use copula framework to estimate dependence between stock returns of a portfolio composed by 94 stock of the S&P100 in order to compute AVaR and VaR and compare the results with respect to a Gaussian Exponentially Weighted Moving Average (EWMA). For computing simulated returns, we use the algorithm presented in [Biglova et all, 2014] and then the model is back-tested with Kupiec's and Christoffersen's tests. The results are coherent with the literature and in particular VaR computed both via copula and EWMA seems to fail to provide an accurate risk measurement while AVaR under copula and EWMA looks more reliable.
Original languageEnglish
Title of host publicationManaging and Modelling of Financial Risks 2016
Subtitle of host publication8th International Scientific Conference
Place of PublicationOstrava
PublisherVSB-Technical University of Ostrava
Pages538-545
Number of pages8
ISBN (Print)9788024839943
Publication statusPublished - 2016
Externally publishedYes
Event8th International Scientific Conference Managing and Modelling of Financial Risks - Ostrava, Czech Republic
Duration: 5 Sept 20166 Sept 2016

Publication series

NameInternational Scientific Conference
ISSN (Print)2464-6973
ISSN (Electronic)2464-6989

Conference

Conference8th International Scientific Conference Managing and Modelling of Financial Risks
Country/TerritoryCzech Republic
CityOstrava
Period5/09/166/09/16

Keywords

  • Value at Risk
  • Average Value at Risk
  • Copula
  • EWMA model

Fingerprint

Dive into the research topics of 'Managing risk with simulated copula'. Together they form a unique fingerprint.

Cite this