Conditional extreme value theory for value at risk and expected shortfall: a comparative backtesting study across financial crises
Ghelfi, Valerio (A.A. 2025/2026) Conditional extreme value theory for value at risk and expected shortfall: a comparative backtesting study across financial crises. Tesi di Laurea in Mathematical finance, Luiss Guido Carli, relatore Sara Biagini, pp. 92. [Bachelor's Degree Thesis]
![[img]](http://tesi.luiss.it/style/images/fileicons/application_pdf.png) |
PDF (Full text)
Restricted to Registered users only Download (2MB) | Request a copy |
Abstract/Index
Why heavy-tailed risk measurement matters. Literature review. The rise and critique of value at risk. Heavy tails: not an anomaly, but a stylized fact. Extreme value theory and the estimation of tail quantiles. From unconditional to conditional EVT. Threshold selection in the pot method. Backtesting risk models. Mathematical framework. Risk measures: value at risk and expected shortfall. Model 1: gaussian parametric. Model 2: historical simulation. Model 3: unconditional EVT. Model 4: conditional EVT. Backtesting theory. Data and methodology. Data sources and sample construction. Crisis windows and evaluation periods. Empirical verification of the stylized facts. POT threshold diagnostics. Rolling-window choices per model. The daily forecasting and backtest pipeline. Reproducibility. In-sample model estimation. Approach. Unconditional EVT: GPD on raw losses. GARCH-EVT step 1: AR (1)-GJR-GARCH (1,1) volatility filter. Residual diagnostics for the GARCH filter. conditional EVT step 2: GPD on standardized innovations. Empirical Results. Rolling risk forecasts: Visual Evidence. VaR backtests: Kupiec and Christoffersen. Expected-Shortfall backtest: Acerbi–Szekely
References
Bibliografia: pp. 80-83.
| Thesis Type: | Bachelor's Degree Thesis |
|---|---|
| Institution: | Luiss Guido Carli |
| Degree Program: | Bachelor's Degree Programs > Bachelor's Degree Program in Economics and Business, English language (L-33) |
| Chair: | Mathematical finance |
| Thesis Supervisor: | Biagini, Sara |
| Academic Year: | 2025/2026 |
| Session: | Summer |
| Deposited by: | Alessandro Perfetti |
| Date Deposited: | 15 Oct 2026 09:40 |
| Last Modified: | 15 Oct 2026 09:40 |
| URI: | https://tesi.luiss.it/id/eprint/46860 |
Downloads
Downloads per month over past year
Repository Staff Only
 |
View Item |
