Confidence intervals of the generalized Pareto distribution parameters based on upper record values (Q2300533)

The paper presented a brief literature review of parameter estimation in the context of the generalized Pareto distribution. The authors presented a new approach to construct confidence intervals for the location and scale parameters of the generalized Pareto distribution, when the shape parameter is known. Furthermore, the authors also did a short simulation study, under several values of the shape parameter and confidence levels, comparing the proposed method with the Edgeworth's approximation method, for the interval estimation of the position and scale parameters of the generalized Pareto distribution. Right after the simulation study, an application is made to real data, more specifically to flood data. The great advantage of the proposed method is that the distributions of pivotal quantities used to build confidence intervals are exact.