- Tytuł:
- Peaks over Threshold Approach with a time-varying scale parameter and range-based volatility estimator for Value-at-Risk and Expected Shortfall estimation
- Autorzy:
- Fałdziński, Marcin
- Tematy:
-
GARCH
Value-at-Risk
Expected Shortfall
Peaks over Threshold
Extreme Value Theory - Pokaż więcej
- Wydawca:
- Główny Urząd Statystyczny
- Powiązania:
- https://bibliotekanauki.pl/articles/59150159.pdf Link otwiera się w nowym oknie
- Opis:
- Exploiting daily high-low range has become increasingly popular among volatility models due to valuable information about volatility dynamics. It has been shown in the literature that range-based volatility estimators can improve volatility and covariance forecasts, and thus models that use high and low prices can outperform standard volatility models based on closing prices solely. This paper incorporates a range-based volatility estimator in an extreme value theory framework to provide better estimates of the tails of daily asset returns. We introduce the Peaks over Threshold model with a range-based volatility estimator depicting the volatility of extreme returns that can contribute to more accurate tail risk estimation. We evaluate the proposed model based on the Monte Carlo simulation and long-period sample of the empirical financial time series by forecasting the Value-at-Risk and Expected Shortfall. We provide evidence that the proposed model can lead to better risk measure forecasts, especially for high tail probabilities.
- Dostawca treści:
- Biblioteka Nauki
Artykuł