A hybrid Pareto model for asymmetric fat-tailed data: the univariate case (Q626276)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | A hybrid Pareto model for asymmetric fat-tailed data: the univariate case |
scientific article |
Statements
A hybrid Pareto model for asymmetric fat-tailed data: the univariate case (English)
0 references
22 February 2011
0 references
The hybrid Pareto distribution (HDP) is built by stitching a generalized Pareto tail to a Gaussian distribution, while enforcing continuity of the resulting density and of its derivatives. Properties of HPDs and maximum likelihood estimates of their parameters are discussed. It is proposed to use mixtures of HDPs for nonparametric density estimation. Results of a simulation study and an application to fire insurance data are presented.
0 references
mixture model
0 references
heavy-tailed distributions
0 references
density estimation
0 references
extreme quantiles
0 references
