Credibility Weighted Hazard Estimation

Credibility weighting is helpful in many insurance applications where sparse data crave information from other sources of data. In this paper we aim at estimating a hazard curve using the nonparametric kernel method, where a credibility weighting principle is used locally, so that areas of sparse da...

Full description

Saved in:
Bibliographic Details
Published in:ASTIN Bulletin : The Journal of the IAA 2000-11, Vol.30 (2), p.405-417
Main Authors: Nielsen, Jens Perch, Sandqvist, Bjørn Lunding
Format: Article
Language:English
Subjects:
Workshop
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Credibility weighting is helpful in many insurance applications where sparse data crave information from other sources of data. In this paper we aim at estimating a hazard curve using the nonparametric kernel method, where a credibility weighting principle is used locally, so that areas of sparse data for one subgroup can be alleviated by available information from other subgroups. The credibility estimator is found through a Hilbert space projection formulation of Buhlmann-Straub's credibility approach.
ISSN:0515-0361
1783-1350
DOI:10.2143/AST.30.2.504643