Treffer: Approximate Bayesian Computations to fit and compare insurance loss models
Title:
Approximate Bayesian Computations to fit and compare insurance loss models
Authors:
Contributors:
Université Claude Bernard Lyon 1 (UCBL), Université de Lyon, Institut de Science Financière et d'Assurances (ISFA), Laboratoire de Sciences Actuarielle et Financière (LSAF), Université de Lyon-Université de Lyon, University of Melbourne
Source:
ISSN: 0167-6687.
Publisher Information:
CCSD
Elsevier
Elsevier
Publication Year:
2021
Collection:
Université de Lyon: HAL
Subject Terms:
likelihood- free inference, risk management, approximate Bayesian computation, Bayesian statistics, [STAT]Statistics [stat], [STAT.AP]Statistics [stat]/Applications [stat.AP], [QFIN.RM]Quantitative Finance [q-fin]/Risk Management [q-fin.RM], [STAT.CO]Statistics [stat]/Computation [stat.CO], [STAT.ME]Statistics [stat]/Methodology [stat.ME]
Document Type:
Fachzeitschrift
article in journal/newspaper
Language:
English
DOI:
10.1016/j.insmatheco.2021.06.002
Availability:
Rights:
info:eu-repo/semantics/OpenAccess
Accession Number:
edsbas.9105DFBE
Database:
BASE
Weitere Informationen
International audience ; Approximate Bayesian Computation (ABC) is a statistical learning technique to calibrate and select models by comparing observed data to simulated data. This technique bypasses the use of the likelihood and requires only the ability to generate synthetic data from the models of interest. We apply ABC to fit and compare insurance loss models using aggregated data. A state-of-the-art ABC implementation in Python is proposed. It uses sequential Monte Carlo to sample from the posterior distribution and the Wasserstein distance to compare the observed and synthetic data. MSC 2010 : 60G55, 60G40, 12E10.