rpot.forecast: Realized Peaks over Threshold forecasts

In lucatrapin/FinEx: Financial econometrics analysis with extreme value statistics

Description

Computes one-day-ahead forecasts of Value at Risk (VaR) and Expected Shortfall (ES) for a fitted RPoT model.

Usage

rpot.forecast(fit, x, alpha)

Arguments

fit	Output object of the rpot() function.
x	Vector of realized measures used in the forecast. The intercept must be included.
alpha	Probability level for the computation of the risk measures.

Details

The risk measures are computed as in Bee et al. (2019):

VaR^{α}=q+\frac{σ}{ξ}≤ft(\frac{φ}{1-α}^{ξ}-1\right)

ES^{α}=\frac{VaR^{α}}{1-ξ}+\frac{σ-qξ}{1-ξ}

Value

A list containing:

parameters

Vector containing the forecasts of phi, sigma, and xi.

var

Var forecast at level alpha.

es

ES forecast at level alpha.

Author(s)

Luca Trapin

References

Bee, M., Dupuis, D. J., and Trapin, L. (2019). Realized Peaks over Threshold: A Time-Varying Extreme Value Approach with High-Frequency-Based Measures. Journal of Financial Econometrics, 17(2), 254-283.

lucatrapin/FinEx documentation built on Dec. 21, 2021, 11:52 a.m.