Alexandre Boumezoued

PhD

Principal

Paris, FR

Alexandre Boumezoued is Principal and Research & Development Director in Milliman’s Paris office.

Experience

Alexandre has led the development of scientific solutions and consulting services over a range of different risk modelling areas, including Economic Scenarios Generators (ESG), asset-related risks, mortality and policyholder behavior, and Economic Capital Models. He has led the expansion of ESG services with the Milliman Economic Scenario Generator (Best ESG Software by Insurance ERM).

He has worked on the following Economic Capital Model projects:

Review and proposals for improvement of the equity risk premium model
Review and implementation of improvements of the sovereign spread risk model
Design of the economic capital model for spread and default risks, including the Dynamic Volatility Adjustment methodology, and their aggregation
Improvement of the Least Square Monte Carlo proxy model validation
Production of benchmark estimates of Internal Model outcomes (market and life risks) for Internal Model and Standard Formula companies
Review and improvement of the mortality, longevity, and pandemic risk models
Review and improvement of the aggregation methodology within different Internal Models
Implementation of a number of improved models for life and financial risks
Review and improvement of the stochastic inflation modelling within the P&C Internal Model for premium and reserve risks
Validation of the reinsurance counterparty default risk and the aggregation methodology for a Partial Internal Model
Review and validation of the market risks modelling and the Nested Simulations Methodology of an Internal Model
Review of the solvency ratio projection process

His consulting experience on ESGs includes:

Implementation and validation of ESGs (Risk Neutral and Real World)
Support to the justification of ESG-embedded parameters to auditors and regulators
Improvement of operational and compliance processes for economic scenarios production
Enhancement of data quality used for ESG calibration
Support for model choices and development

He has worked on the following projects related to mortality modeling:

Enhancing the data quality for general population mortality estimates
Modelling the impact of heat waves on mortality
Implementing SIRD-type models for pandemic severity forecasting
Implementing a number of improved models for longevity, mortality, and mortality CAT risks
Retreating pandemic mortality within mortality series and other historical patterns
Analysing cause-of-death mortality differentials by socio-economic groups for the Society of Actuaries
Reviewing frequency and severity components of a pandemic model, including the spread mechanisms within the spatial model, analysis of model sensitivity, and proposal of modeling simplifications
Deriving mortality and longevity scenarios for updating Best Estimate assumptions used in insurance valuation and risk transfer

As part of his scientific activity, Alexandre is lecturer in actuarial centers in France, and referee for scientific journals.

Publications and Presentations

Publications

Boumezoued, A., Cherkaoui, Y., & Hillairet, C. 2023. Cyber risk modeling using a two-phase Hawkes process with external excitation.

Andres, H., Bourmezoued, A., & Jourdain, B. 2022. Signature-based validation of real-world economic scenarios.

Boumezoued, A. 2021. Discussion on paper “Shapovalov et al (2021), Exchangeable mortality projection.” European Actuarial Journal.

Boumezoued, A. & Lecoeur, E. 2021. Actuarial R&D – why do it, and how? The Actuary, April 2021.

Boumezoued, A. & Robert, C. 2020. Individual claim reserving models: parametric vs non-parametric models. In the book: Insurance Data Analytics.

Boumezoued, A. & Elfassihi, A. 2020. Mortality data correction in the absence of monthly fertility records. Insurance: Mathematics and Economics.

Boumezoued, A., & Lecoeur, E. 2020. Existence et conscience de la R&D Actuarielle. L'Actuariel no 30, Octobre, 2020.

Andres, H., Arrouy, P.E., Bonnefoy, P., Boumezoued, A., & Mehalla, S. 2020. Fast calibration of the LIBOR Market Model with Stochastic Volatility based on analytical gradient.

COVID-19 outbreak: what can actuaries learn about mortality? 2020. The European Actuary.

Arrouy, P. E., Boumezoued, A., Lapeyre, B., & Mehalla, S. 2022. Jacobi stochastic volatility factor for the LIBOR market model. Finance and Stochastics, 26(4), 771-823.

Bessy-Roland, Y., Boumezoued, A., & Hillairet, C. 2021. Multivariate Hawkes process for cyber insurance. Annals of Actuarial Science, 15(1), 14-39.

Boumezoued, A., Hoffmann, M., & Jeunesse, P. 2021. Nonparametric adaptive inference of birth and death models in a large population limit. Mathematical Staticstics and Learning, 3(1), 1-69.

Boumezoued, A., Hoffmann, M., & Jeunesse, P. 2018. A new inference strategy for general population mortality tables. ASTIN Bulletin. https://hal.archives-ouvertes.fr/hal-017773665.

Balland, F., Boumezoued, A., Devineau, L., Habart, M., & Popa, T. 2018. Mortality data reliability in an internal model. Annals of Actuarial Science.

Devineau, L., Arrouy, P.E., Bonnefoy, P., & Boumezoued, A. 2017. Fast calibration of the LIBOR Market Model with Stochastic Volatility and Displaced Diffusion. Journal of Industrial and Management Optimization.

Boumezoued, A. & Devineau, L. 2017. Individual claims reserving: a survey.

Boumezoued, A. 2016. Population viewpoint on Hawkes processes. Advances in Applied Probability, Volume 48, Pages 1-18.

Boumezoued, A. 2016. Improving HMD mortality estimates with HFD fertility data. North American Actuarial Journal.

Arnold, S., Boumezoued, A., Labit Hardy, H., & El Karoui, N. 2016. Cause-of-Death Mortality: What Can Be Learned From Population Dynamics?. Insurance: Mathematics and Economics (Longevity 11 Special Issue).

Boumezoued, A., El Karoui, N., & Loisel, S. 2015. Measuring mortality heterogeneity dynamics with internal-censored data. Insurance: Mathematics and Economics. Volume 72, January 2017, Pages 67-82.

Boumezoued, A. 2016. Micro-macro analysis of heterogenous age-structured populations dynamics. Application to self-exciting processes and demography. Doctoral thesis.

Boumezoued, A. 2015. Macroscopic behavior of heterogenous populations with fast random life histories.

Boumezoued, A., Angoua, Y., Devineau, L., & Boisseau, J.P. 2011. One-year reserve risk including a tail factor: closed formula and bootstrap approaches.

Professional Designations

Professional Fellow of Louis Bachelier Institute

Education

Engineer (Ecole Nationale des Ponts et Chaussées)
MSc in Probability Theory (Paris 6 University)
PhD in Applied Mathematics (Probability and Random Models Laboratory, Paris 6 University)

Publications

Read their latest work

Video

Harnessing AI for insurance: Real-world success stories

13 February 2025 - by Alexandre Boumezoued, Paul Marron

The Milliman AI team has developed solutions in the areas of automating claims processing, chatbot development, model validation, and report generation.

Article

Backtesting of dynamic hedging of Registered Indexed-Linked Annuity (RILA)

04 December 2024 - by Hervé Andrès, Alexandre Boumezoued, Ken Qian, Katherine Wang

We show how dynamic hedging of registered indexed-linked annuities can often reasonably predict resulting profit or loss.

Article

Modeling financial losses from a ransomware attack using a causal approach

13 October 2023 - by Chris Beck, Alexandre Boumezoued, Yousra Cherkaoui Tangi, Eliott Pradat

With ransomware attacks on the rise all over the world, regulators are keeping a close eye on the reimbursement of ransoms by insurers.

Article

An epidemiological model for silent cyber accumulation risk assessment

09 March 2023 - by Alexandre Boumezoued, Yousra Cherkaoui Tangi, Thomas Peyrat

We present a toy portfolio inspired by OECD sectors data where the interactions between the insured are modeled through a network.

Article

Modeling the impact of climate risks on mortality

19 December 2022 - by Alexandre Boumezoued, Amal Elfassihi, Valentin Germain, Eve Titon

Climate change will lead to greater mortality risk, so we discuss the exposure to high temperatures and look at alternative modeling for vector-borne diseases.

Article

A new hybrid Random Number Generator for more accurate valuation of insurance liabilities

12 December 2022 - by Hervé Andrès, Pierre-Edouard Arrouy, Paul Bonnefoy, Alexandre Boumezoued

Increasing use of stochastic economic scenarios for valuation of liabilities has put more pressure on the operational process of carriers, but the RNG can help.

Article

Calibration accuracy of three variants of the Libor Market Model

01 December 2022 - by Pierre-Edouard Arrouy, Paul Bonnefoy, Elias Bouiti, Alexandre Boumezoued, Julien Vedani

We highlight a Libor market model with constant elastic volatility, showing an interesting trade-off between parameters used and quality of results.

Article

Mortality trend prediction using machine learning

23 November 2022 - by Amal Elfassihi, Alexandre Boumezoued

Compared with traditional mortality models, machine learning algorithms can significantly improve the forecasts of future mortality rates.

Article

Efficient computation of Solvency Capital Requirement using Multilevel Monte Carlo Methods

08 November 2022 - by Alexandre Boumezoued, Adel Cherchali, Mathieu Truc

MLMC methods represent a relevant alternative to the proxy modelling approaches commonly used in capital requirement estimation within the insurance sector.

Article

The impact of carbon risk factor on equity dynamics

14 October 2022 - by Alexandre Boumezoued, Sophian Mehalla, Valentin Germain

With better understanding of climate transition risk exposure and stress tests proposed by regulators, internal models can help provide insights.