"sendo l'intento mio scrivere cosa utile a chi la intende..."
Earlier this year, I was reading Arnon et al.’s Science paper, “Whale song shows language-like statistical structure,” which reports statistical regularities in humpback whale song that resemble those found in human language, including a frequency distribution close to Zipf’s law. It brought me back to a simple question: even if we observe a Zipf law, what can we really infer about the mechanism…
Motor insurance models and road-safety studies address the same underlying risk, but at very different scales. Insurance models typically predict annual claim counts from a small set of rating variables, such as age, mileage, vehicle characteristics, or past claims. Road-safety research, by contrast, seeks to understand mechanisms operating much closer to the crash itself, including speed,…
(avec le sous-titre De l’actuariat contemplatif à l’actuariat prédictif ) En mars 2024, Gavin Schmidt, directeur du Goddard Institute for Space Studies de la NASA, publiait dans Nature un texte au titre inhabituellement inquiet : Climate models can’t explain 2023’s huge heat anomaly — we could be in uncharted territory. L’année 2023 avait surpris les climatologues. Plusieurs facteurs connus…
My short article, when urban flood models disagree: hydrological equifinality as actuarial model risk, was recently published in the European Actuarial Journal. In a nutshell, I tried to address the question how much does a flood that has not happened yet cost? For insurers, answering that question involves a surprisingly long chain: rainfall has to become runoff, runoff has to become water in…
We still had one last day of work with Marie-Pier and Olivier—hopefully with something to show for it by the start of the academic year?—before taking a few days off. At least, I think so. I hope so.
Après plus de dix ans à écrire des billets de blog et des textes destinés à un public plus large, j’ai profité de cette année pour reprendre, actualiser et rassembler une partie de ces réflexions. Le résultat est un petit livre non technique, La Société au prix du risque — Assurance, algorithmes et climat à l’âge de la prédiction. À mesure que les données deviennent plus fines et les modèles plus…
After more than ten years of writing blog posts and essays for a broader audience, I used this past year to revisit, update, and bring some of those ideas together. The result is a short, non-technical book, Risk-Priced Society — Insurance, Algorithms and Climate in the Age of Prediction. As data become more granular and models more powerful, insurers can distinguish more precisely between people,…
L article La technologie nous sauvera-t-elle ? a été publié dans le dernier numéro de Risques.
Source: a surreal and obsessive entertainment conceived and produced in 1992 for the birth of ARTE channel, with the aim of improving the quality of life of insomniacs and enlivening the nights of night owls, by Hélène Guétary, soundtrack Nova/Andrew Orr
L été prochain, on propose 3 stages, à Montréal. Modeling Prevention in Insurance Economics Transfer Learning for Extreme Events Computational Optimal Transport Le site de https://www.mitacs.ca/fr-ca/ va ouvrir dans les jours à venir (oui, il faut s y prendre très en avance). Pour faire simple on a proposé des sujets les étudiants postulent sur la plateforme de mitacs à la fin de l été cet…
After some pictures from Kyoto 京都, Fall and Winter, some pictures from Spring and Summer. In two weeks, my sabbatical year at Kyoto University will come to an end. It is difficult to sum up what this year has meant. It gave me the time and space to look at things differently, to revisit projects that needed a fresh perspective, and to let new ideas emerge. But above all, it was a wonderful…
Before leaving Kyoto to go back to Montréal (and enjoy some summer break), I uploaded some lecture notes, on Optimal Transport for Actuarial Science, on HAL (soon on ArXiv ?). These lecture notes introduce optimal transport as a mathematical language for actuarial science. They treat losses, premiums, scores, reserves, capital scenarios, climate losses and lifetime distributions as probability…
Yesterday, Jonno White published a a list of 50 actuarial thought leaders who genuinely deserve to be far better known outside the profession itself . I have to admit that I am proud, honoured (and a little surprised) to be included in this list, alongside colleagues whose work I deeply admire (such as Moshe Milevsky, David Blake, Andrew Cairns, Johnny Siu-hang Li, Gordon Woo, Ron Richman, Ioannis…
In The Actuary, the official magazine of the Institute and Faculty of Actuaries, in the UK, a short article. AI is often discussed as a tool for insurers: a way to improve pricing, underwriting, fraud detection, claims management and customer service. But it is also becoming an object of insurance. Firms use AI systems in customer service, legal drafting, software development, marketing, finance,…
The 29th International Congress on Insurance: Mathematics and Economics (IME 2026) will be held in Seoul, I will present on Friday our joint work with Bertille Tierny and François Hu, Direct and Indirect Discrimination in Generalized Linear Models (the paper is available on arxiv) Generalized linear models are central to actuarial modelling of binary risk, claim frequency, utilization, and…
This afternoon, I will give a series of two short lectures on fairness and performance in actuarial modeling. Slides are now available.
In three weeks, the International Workshop on Risk and Insurance will be organized in 서울 (Seoul, Korea), prior to the Insurance: Mathematics Economics conference. We were asked to send our slides in advance, so I thought I could be nice to share them (slides are available here). I will be the very first speaker of the day, in a session AI Revolution and Cyber Risks . Over the past ten years, we…
J’ai déjà parlé, ici ou là, de mes souvenirs d’enfance et des cours de français au secondaire. En particulier de mon incompréhension assez profonde devant l’analyse de texte littéraire, où il fallait, me semblait-il, deviner les intentions de l’auteur, repérer des symboles que je ne voyais pas, attribuer à une tournure ou à une image une profondeur que je soupçonnais parfois d’être reconstruite…
Our article, Recoverability of market-wide fair insurance premiums under selection bias, with Marie-Pier Côté and Olivier Côté, was published in Insurance: Mathematics Economics. Fairness adjustments in insurance pricing are defined relative to a reference population, i.e., to a joint distribution of (X, D, Y) where X are rating factors, D protected attributes, and Y is claim amount. Because an…
Last week, the paper Perceived fairness in networks was published in Network Science. The usual definitions of algorithmic fairness focus on population-level statistics, such as demographic parity or equal opportunity. However, in many social or economic contexts, fairness is not perceived globally, but locally, through an individual’s peer network and comparisons. We propose a theoretical model…
Il y a deux jours, je revenais sur une (ancienne) étude de l’Ifop pour Fairgen sur l’intention de vote des enseignants aux élections européennes. La note méthodologique expliquant que l’enquête avait été menée auprès de 8 000 personnes représentatives de la population française adulte, “incluant l’équivalent statistique de 580 enseignants de collège et de lycée”, ces 580 enseignants étant “issus…
Au Québec, on utilise une jolie expression, « pelleter en avant », ou « pelleter par en avant », Fait référence à quelqu un qui déneige en pelletant la neige en avant de lui, ce qui l amène à cumuler cette dernière dans son chemin. Il doit donc déplacer une quantité de plus en plus grande de neige au fur et à mesure qu il progresse. Cela fait en sorte, notamment, qu un jour ou l autre, il ne…
J’ai découvert un peu par hasard (via un post sur bsk de Thomas Delclite) une étude de l’Ifop pour Fairgen sur l’intention de vote des enseignants aux élections européennes. La note méthodologique indique que l’enquête a été menée auprès de 8 000 personnes représentatives de la population française adulte, “incluant l’équivalent statistique de 580 enseignants de collège et de lycée”, ces 580…
Notre tribune, Risques climatiques : la donnée et l’intelligence artificielle ne remplaceront pas la solidarité, écrite avec Laurence Barry, a été publiée par L Argus de l Assurance. À mesure que les risques climatiques s’intensifient, l’essor des données fines et de l’intelligence artificielle promet une meilleure connaissance des expositions. Mais derrière cette avancée technique se joue un…
Tomorrow, Raphaël Dalbarade will attend the European Geosciences Union (EGU) Annual Meeting, in Vienna, to present When Climate Hazard Granularity Challenges Risk Pooling: A Spatial Perspective , a joint work with Laurence Barry, Caroline Hillairet, Hamza El Hassani, Azeddine Bamansor, Quentin Hénaff, and Simon Blaquière. More very soon The French CatNat regime provides mandatory natural disaster…
Two years ago, my book Insurance, Biases, Discrimination and Fairness was published in the Springer Actuarial series. Discrimination in insurance is a difficult topic because, in a very specific sense, insurers are expected to discriminate: they classify risks, build risk pools, and differentiate premiums. This is the logic of risk-based pricing and actuarial fairness. But insurance is not only…
Un rapport présentant l assurance des catastrophes naturelles au Japon est en ligne dans la collection des documents de la Chaire PARI, Cette note montre que le Japon a construit, face à une exposition exceptionnelle aux séismes, tsunamis, volcans, inondations et glissements de terrain, un système de couverture des catastrophes naturelles original, qui ne repose ni sur un régime unifié de type…
I was invited to give a talk at the IFoA AI Ethics Governance Risk Management forum at the end of the week. I uploaded some slides to start the discussion. I will not give a highly technical lecture, but rather to propose a way of thinking about fairness in insurance AI that may be useful for actuaries, for governance people, and for anyone involved in model risk or oversight. The main message is…
This post was initially written in French, La technologie nous sauvera-t-elle ? I feel as though I keep hearing, more and more often, that climate change is above all a problem of innovation. Emissions continue to rise, targets keep slipping out of reach, yet we fill our collective imagination with carbon-capture machines, artificial intelligences supposedly able to optimize the transition, and…
J ai l impression d entendre de plus en plus souvent dire que le climat est avant tout un problème d’innovation. Les émissions continuent d’augmenter, les objectifs se dérobent, mais on essaye de peupler notre imaginaire collectif de machines à capturer le carbone, d’intelligences artificielles capables d’optimiser la transition, voire de techniques de modification du climat. Cette confiance n’a…