GParetoVaR function

VaR for Generalized Pareto

Estimates the Value at Risk of a portfolio assuming losses are distributed as a generalised Pareto.

GParetoVaR(Ra, beta, zeta, threshold.prob, cl)

Arguments

• Ra: Vector of daily Profit/Loss data
• beta: Assumed scale parameter
• zeta: Assumed tail index
• threshold.prob: Threshold probability corresponding to threshold u and x
• cl: VaR confidence level

Returns

Expected Shortfall

Examples

# Computes ES assuming generalised Pareto for following parameters
   Ra <- 5 * rnorm(100)
   beta <- 1.2
   zeta <- 1.6
   threshold.prob <- .85
   cl <- .99
   GParetoVaR(Ra, beta, zeta, threshold.prob, cl)

Author(s)

Dinesh Acharya

References

Dowd, K. Measuring Market Risk, Wiley, 2007.

McNeil, A., Extreme value theory for risk managers. Mimeo, ETHZ, 1999.