Nonparametric density estimation and risk quantification from tabulated sample moments

Philippe Lambert

No 2023001, LIDAM Reprints ISBA from Université catholique de Louvain, Institute of Statistics, Biostatistics and Actuarial Sciences (ISBA)

Abstract: Continuous data such as losses are often summarized by means of histograms or displayed in tabular format: the range of data is partitioned into consecutive interval classes and the number of observations falling within each class is provided to the analyst. This paper investigates how the additional report of sample moments within each class can be integrated to obtain a smooth nonparametric estimate of the density and credible intervals for the loss quantiles. Extensive simulations confirm the merits of the proposed methodology with correctly estimated densities based on tabulated sample moments of increasing orders and effective coverages of credible intervals close to their nominal values, even when the number of classes is small. An application on motor insurance data further illustrates the usefulness of the method with an estimation of the loss density and of Value-at-Risk.

Keywords: Nonparametric density estimation; Grouped data; Tabulated sample moments; Value-at-Risk; P-splines (search for similar items in EconPapers)
JEL-codes: C10 C14 C18 (search for similar items in EconPapers)
Pages: 20
Date: 2023-01-01
Note: In: Insurance: Mathematics and Economics, 2023, vol. 108, p. 177-189
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:aiz:louvar:2023001

DOI: 10.1016/j.insmatheco.2022.12.004

Access Statistics for this paper

More papers in LIDAM Reprints ISBA from Université catholique de Louvain, Institute of Statistics, Biostatistics and Actuarial Sciences (ISBA) Voie du Roman Pays 20, 1348 Louvain-la-Neuve (Belgium). Contact information at EDIRC.
Bibliographic data for series maintained by Nadja Peiffer ().