Bagging and regression trees in individual claims reserving
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F25%3A10498137" target="_blank" >RIV/00216208:11320/25:10498137 - isvavai.cz</a>
Result on the web
<a href="https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=pxv_.PYU0V" target="_blank" >https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=pxv_.PYU0V</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/s00362-025-01715-9" target="_blank" >10.1007/s00362-025-01715-9</a>
Alternative languages
Result language
angličtina
Original language name
Bagging and regression trees in individual claims reserving
Original language description
This methodological paper presents a novel approach to individual claims reserving in non-life insurance, utilizing machine learning techniques. Claims reserving in insurance amounts to stochastically predict the overall loss reserves to cover possible future claims. The developed concepts leverage regression trees and bootstrap aggregating (bagging) to improve the accuracy of reserve predictions. Unlike current approaches focusing solely on the number of claims so far, our approach models both the frequency and severity of claims. Out-of-bag error is employed as a diagnostic tool to enhance model validation. The effectiveness of the proposed methodology is demonstrated through an exemplary data analysis, showcasing its potential to provide more accurate reserve estimates in claims reserving.
Czech name
—
Czech description
—
Classification
Type
J<sub>imp</sub> - Article in a specialist periodical, which is included in the Web of Science database
CEP classification
—
OECD FORD branch
10103 - Statistics and probability
Result continuities
Project
<a href="/en/project/GF23-06461K" target="_blank" >GF23-06461K: Sparse solutions of underdetermined systems: retroactive applications, algorithms, inference, and new perspectives</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2025
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Name of the periodical
Statistical Papers
ISSN
0932-5026
e-ISSN
1613-9798
Volume of the periodical
66
Issue of the periodical within the volume
4
Country of publishing house
DE - GERMANY
Number of pages
26
Pages from-to
89
UT code for WoS article
001482788000003
EID of the result in the Scopus database
2-s2.0-105004357847