ALESSANDRO GUSTAVO LAPORTA

Dottore di ricerca

ciclo: XXXIV



Titolo della tesi: A Neural Network approach to measure health insurance risk

This work presents a set of neural network applications to health insurance pricing. In recent years, the actuarial literature involving machine learning in insurance pricing has flourished. However, most actuarial machine learning research focuses on car and property and casualty insurance. While, the use of such techniques in health insurance is yet to be explored. In this manuscript, we discuss the use of neural networks to set the price of an health insurance coverage following the structure of a classical frequency-severity model. We consider neural networks to estimate claim frequency and severity. In particular, we introduce Negative Multinomial Neural Networks to jointly model the frequency of possibly correlated medical claims. We then complete the frequency-severity approach proposing Gamma Neural Networks to estimate the expected claim severity. We then go beyond the frequency-severity framework adopting a quantile approach that allows gauging the potential riskiness of a given policyholder. Namely, we discuss the estimation of conditional quantiles of aggregate claim amounts embedding the problem in a quantile regression framework using the Neural Network approach. As the first step, we consider Quantile Regression Neural Networks (QRNN) to compute quantiles for the insurance ratemaking framework. As the second step, we propose a new Quantile Regression Combined Actuarial Neural Network (Quantile-CANN) combining the traditional quantile regression approach with a Quantile Regression Neural Network. In both cases, we adopt a two-part model scheme where we fit a logistic regression to estimate the probability of positive claims and the QRNN model or the Quantile-CANN for the positive outcomes. Through a case study based on a health insurance dataset, we highlight the overall better performances of the different neural network models with respect to more established regression models (such as GLMs and quantile regression), both in terms of accuracy and risk diversification.

Produzione scientifica

11573/1682197 - 2024 - Neural Networks for quantile claim amount estimation: a quantile regression approach
Laporta, Alessandro G.; Levantesi, Susanna; Petrella, Lea - 01a Articolo in rivista
rivista: ANNALS OF ACTUARIAL SCIENCE (- Cambridge: Cambridge University Press -[London]: Published by the Institute of Actuaries and the Faculty of Actuaries, 2006-) pp. 30-50 - issn: 1748-5002 - wos: WOS:001007747300001 (2) - scopus: (0)

11573/1604641 - 2021 - Quantile Regression Neural Network for Quantile Claim Amount Estimation
Laporta, Alessandro G.; Levantesi, Susanna; Petrella, Lea - 04b Atto di convegno in volume
congresso: Mathematical and Statistical Methods for Actuarial Sciences and Finance: eMAF2020 (on-line)
libro: Mathematical and Statistical Methods for Actuarial Sciences and Finance: eMAF2020 - (978-3-030-78964-0; 978-3-030-78965-7)

11573/1185689 - 2018 - Selection of Value at Risk Models for Energy Commodities
Laporta, Alessandro Gustavo; Merlo, Luca; Petrella, Lea - 01a Articolo in rivista
rivista: ENERGY ECONOMICS ([S.l.] : Elsevier Science) pp. 628-643 - issn: 1873-6181 - wos: WOS:000445713500044 (53) - scopus: 2-s2.0-85050521957 (60)

11573/1115365 - 2018 - Cross‑Country assessment of systemic risk in the European Stock Market: evidence from a CoVaR analysis
Petrella, Lea; Laporta, Alessandro Gustavo; Merlo, Luca - 01a Articolo in rivista
rivista: SOCIAL INDICATORS RESEARCH (Dordrecht; London; Boston: Springer Nature B.V. Dordrecht; London; Boston: Kluwer Academic Publishers; Reidel) pp. - - issn: 0303-8300 - wos: WOS:000505945700011 (13) - scopus: 2-s2.0-85044177337 (11)

© Università degli Studi di Roma "La Sapienza" - Piazzale Aldo Moro 5, 00185 Roma