Numerical Solution to the Problem of Optimal Boundary of Generalized Pareto Mixtures
In this paper we introduce a solution to the optimal boundary of a mixture of a generalized Pareto distribution
when an explicit boundary is not available. The solution is based on Newton-Raphson iteration method. A special attention
is paid to two-component mixtures of equal scale and shape parameters. Three Generalized Pareto sub-models are
considered, namely, exponential, Pareto and beta as well as the unified model.
Keywords— Bayesian Risk, Discriminant Function, Optimal Boundary, Pareto Mixture, Newton-Raphson Method.