Value-at-Risk Estimation: A Copula-GARCH Approach
Author
Sørensen, Kirstine Lykke
Term
4. term
Education
Publication year
2023
Submitted on
2023-06-02
Pages
79
Abstract
Specialet analyserer Value-at-Risk (VaR) for en portefølje bestående af de 10 største aktiver i S&P 500-indekset. VaR er et udbredt mål for, hvor stort et tab man med en given sandsynlighed kan opleve over en kort periode. Porteføljen modelleres med en copula-GARCH-model, der både beskriver udsvingene i de enkelte aktiver (GARCH) og afhængigheden mellem dem (copula). Risikoen estimeres ved hjælp af Monte Carlo-estimater, dvs. mange computersimuleringer baseret på tilfældige træk. Endelig sammenlignes resultaterne med den parametriske VaR for selve S&P 500-indekset.
This thesis analyzes the Value-at-Risk (VaR) of a portfolio made up of the 10 largest assets in the S&P 500 index. VaR is a widely used measure of how large a loss one might face over a short period with a given probability. The portfolio is modeled using a copula-GARCH approach, which captures both the volatility of individual assets (GARCH) and the dependence between them (copula). Risk is evaluated using Monte Carlo estimates, that is, many computer simulations based on repeated random draws. Finally, the results are compared with the parametric VaR of the S&P 500 index.
[This summary has been rewritten with the help of AI based on the project's original abstract]
Keywords
copula ; ARMA-GARCH ; Value-at-Risk ; S&P500
Documents