09:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
10:00 - 10:30 (30min)
Accueil et Café
10:30 - 11:15 (45min)
Quentin GUIBERT
Salle des conférences-IRA
-Title : Impacts of Climate Change on Mortality: An extrapolation of temperature effects based on time series data in France.
- Abstract :Most contemporary mortality models rely on extrapolating trends or past events. However, population dynamics will be significantly impacted by climate change, notably the influence of temperatures on mortality. In this paper, we introduce a novel approach to incorporate temperature effects on projected mortality using a multi-population mortality model. This method combines a stochastic mortality model with a climate epidemiology model, predicting mortality variations due to daily temperature fluctuations, be it excesses or insufficiencies. The significance of this approach lies in its ability to disrupt mortality projections by utilizing temperature forecasts from climate models and to assess the impact of this unaccounted risk factor in conventional mortality models.
- Abstract :Most contemporary mortality models rely on extrapolating trends or past events. However, population dynamics will be significantly impacted by climate change, notably the influence of temperatures on mortality. In this paper, we introduce a novel approach to incorporate temperature effects on projected mortality using a multi-population mortality model. This method combines a stochastic mortality model with a climate epidemiology model, predicting mortality variations due to daily temperature fluctuations, be it excesses or insufficiencies. The significance of this approach lies in its ability to disrupt mortality projections by utilizing temperature forecasts from climate models and to assess the impact of this unaccounted risk factor in conventional mortality models.
11:15 - 12:00 (45min)
Marie-Pierre ETIENNE
Salle des conférences-IRA
Title: Des équations différentielles stochastiques pour le suivi des espèces sauvages.
- Abstract : L’écologie a pour but d’étudier les rapports que les êtres vivants entretiennent avec leur milieu. Les techniques d’observation permettant de mieux comprendre ces rapports ont connu ces derniers années un développement sans précédent, suivant notamment la généralisation de l’utilisation de capteurs. En particulier le développement de système de suivi GPS léger et très autonome permet de suivre les déplacements des individus de manière précise et peu invasive. La compréhension du déplacement permet d’étudier la répartition spatiale des animaux sauvages, pour mieux comprendre le lien avec les caractéristiques environnementales et par suite d’aborder des questions de conservation. De nombreuses méthodes statistiques, issues du traitement du signal, propose d’analyser les données de déplacement à l’aide de modèle à temps discret. Toutefois, les modèles à temps continu présenten
- Abstract : L’écologie a pour but d’étudier les rapports que les êtres vivants entretiennent avec leur milieu. Les techniques d’observation permettant de mieux comprendre ces rapports ont connu ces derniers années un développement sans précédent, suivant notamment la généralisation de l’utilisation de capteurs. En particulier le développement de système de suivi GPS léger et très autonome permet de suivre les déplacements des individus de manière précise et peu invasive. La compréhension du déplacement permet d’étudier la répartition spatiale des animaux sauvages, pour mieux comprendre le lien avec les caractéristiques environnementales et par suite d’aborder des questions de conservation. De nombreuses méthodes statistiques, issues du traitement du signal, propose d’analyser les données de déplacement à l’aide de modèle à temps discret. Toutefois, les modèles à temps continu présenten
12:00 - 13:30 (1h30)
Déjeuner
13:30 - 14:15 (45min)
Sarah KAAKAI
Salle des conférences-IRA
-Title: Estimation of Systemic Shortfall Risk Measure using Stochastic Algorithms.
- Abstract :Systemic risk measures are multivariate risk measures, designed to capture both the global risk of an interconnected system and the individual risk associated with each component of the system. In this talk, I will focus on multivariate systemic shortfall risk measures (MSRM), which can be interpreted as the minimal amount of cash that secures the system by allocating capital to each single component before aggregating individual risks. While the theoretical foundation of these risk measures has been well explored by various authors, the numerical aspects have received less attention. I will present stochastic algorithm schemes for the estimation of MSRM, and prove that the resulting estimators are consistent and asymptotically normal. Additionally, I will share numerical results that test the performance of these algorithms across several examples. This is a joint work with
- Abstract :Systemic risk measures are multivariate risk measures, designed to capture both the global risk of an interconnected system and the individual risk associated with each component of the system. In this talk, I will focus on multivariate systemic shortfall risk measures (MSRM), which can be interpreted as the minimal amount of cash that secures the system by allocating capital to each single component before aggregating individual risks. While the theoretical foundation of these risk measures has been well explored by various authors, the numerical aspects have received less attention. I will present stochastic algorithm schemes for the estimation of MSRM, and prove that the resulting estimators are consistent and asymptotically normal. Additionally, I will share numerical results that test the performance of these algorithms across several examples. This is a joint work with
14:15 - 15:00 (45min)
Emmanuel LEPINETTE
Salle des conférences-IRA
-Title : Pricing without martingale measure: an alternative approach for general (non linear) models.
15:00 - 15:15 (15min)
Pause café
15:15 - 16:00 (45min)
Stéphane LOISEL
Salle des conférences-IRA
-Title : Quelques idées de collaboration sur la thématique de la prévention: de l'actuariat à la biodiversité.
-Abstract : Dans cet exposé nous présentons des résultats obtenus sur les stratégies de prévention optimale obtenues dans un cadre de modélisation actuarielle classique (la théorie de la ruine) et décrivons quelques problématiques liées à la bioversité qui pourraient être abordées avec des membres de Matrisk.
-Abstract : Dans cet exposé nous présentons des résultats obtenus sur les stratégies de prévention optimale obtenues dans un cadre de modélisation actuarielle classique (la théorie de la ruine) et décrivons quelques problématiques liées à la bioversité qui pourraient être abordées avec des membres de Matrisk.
16:00 - 16:45 (45min)
Solym MANOU ABI
Salle des conférences-IRA
-Title : Stable driven stochastic models with financial applications.
-Abstract: In this talk, I am interested in stable driven stochastic models that are often used in finance, either to compute option prices using stochastic orders (which are decision tools in financial risk theory) or to estimate the associated parameters of these models.
-Abstract: In this talk, I am interested in stable driven stochastic models that are often used in finance, either to compute option prices using stochastic orders (which are decision tools in financial risk theory) or to estimate the associated parameters of these models.
19:30 - 22:00 (2h30)
Dîner au restaurant "Le Nez Rouge"
9:30 - 10:15 (45min)
Peter TANKOV
Salle des conférences-IRA
-Title : Asset pricing under transition scenario uncertainty and model ambiguity.
- Abstract:We study asset pricing and optimal investment decisions for an economic agent whose future revenues depend on the realization of a scenario from a given set of possible futures. In the first part of the talk, we assume that future scenario is unknown, but that the possible scenarios have equal probabilities, and the agent deduces scenario information progressively by observing a signal. The problem of valuing an investment is formulated as an American option pricing problem with Bayesian learning. In the second part, we assume that the probabilities of individual prospective scenarios are ambiguous and place ourselves into the smooth model of decision making under ambiguity aversion of Klibanoff et al (2005), framing the optimal investment decision as an optimal stopping problem with learning under
- Abstract:We study asset pricing and optimal investment decisions for an economic agent whose future revenues depend on the realization of a scenario from a given set of possible futures. In the first part of the talk, we assume that future scenario is unknown, but that the possible scenarios have equal probabilities, and the agent deduces scenario information progressively by observing a signal. The problem of valuing an investment is formulated as an American option pricing problem with Bayesian learning. In the second part, we assume that the probabilities of individual prospective scenarios are ambiguous and place ourselves into the smooth model of decision making under ambiguity aversion of Klibanoff et al (2005), framing the optimal investment decision as an optimal stopping problem with learning under
10:15 - 11:00 (45min)
Sara BIAGINI
Salle des conférences-IRA
-Title: Carbon neutrality and net-zero in compliance markets.
- Abstract: When addressing climate risk mitigation, a primary objective is achieving climate neutrality. We analyze first the impact of carbon neutrality policies on a system of polluting companies that are regulated within an ETS carbon market. The companies are mandated to cut emissions and have the option to trade carbon allowances for compliance purposes. They may also be allocated additional allowances as subsidies from the regulatory authority to assist in fulfilling compliance. For each firm, the key variable is the imbalance between their cumulative Business As Usual emissions and the allocation they may receive, which must be covered by the regulated maturity. For a given subsidies scheme, a unique, closed-form equilibrium is reached. The resulting allowances price admits a neat expression as a convex combination of each company’s marginal costs, each calculated based on their predicted emi
- Abstract: When addressing climate risk mitigation, a primary objective is achieving climate neutrality. We analyze first the impact of carbon neutrality policies on a system of polluting companies that are regulated within an ETS carbon market. The companies are mandated to cut emissions and have the option to trade carbon allowances for compliance purposes. They may also be allocated additional allowances as subsidies from the regulatory authority to assist in fulfilling compliance. For each firm, the key variable is the imbalance between their cumulative Business As Usual emissions and the allocation they may receive, which must be covered by the regulated maturity. For a given subsidies scheme, a unique, closed-form equilibrium is reached. The resulting allowances price admits a neat expression as a convex combination of each company’s marginal costs, each calculated based on their predicted emi
11:00 - 11:15 (15min)
Pause café
11:15 - 12:00 (45min)
José GARRIDO
Salle des conférences-IRA
Title: The French Actuarial Climate Index and its Application to Parametric Insurance.
-Abstract:Climate change is often defined as the long–term fluctuations in climate patterns affecting the planet globally. Among its main observed effects are a rise in average temperatures in many parts of the globe, and an increase in the frequency and severity of extreme weather events, such as floods, droughts, or windstorms. These new climate risks are increasingly affecting the frequency and the severity of claims in different insurance branches. In order to help insurance companies predict and manage climate risks in North America, local actuaries have created the Actuaries Climate IndexTM (ACI). Here we consider the extension of the North American ACI methodology to climate data from France, including its largest island, Corsica. We calculate the French Actuarial Climate Index (FACI) with data extracted from the ERA5–Land reanalysis database, see [1]. Unlike the ACI for the United S
-Abstract:Climate change is often defined as the long–term fluctuations in climate patterns affecting the planet globally. Among its main observed effects are a rise in average temperatures in many parts of the globe, and an increase in the frequency and severity of extreme weather events, such as floods, droughts, or windstorms. These new climate risks are increasingly affecting the frequency and the severity of claims in different insurance branches. In order to help insurance companies predict and manage climate risks in North America, local actuaries have created the Actuaries Climate IndexTM (ACI). Here we consider the extension of the North American ACI methodology to climate data from France, including its largest island, Corsica. We calculate the French Actuarial Climate Index (FACI) with data extracted from the ERA5–Land reanalysis database, see [1]. Unlike the ACI for the United S
12:00 - 12:45 (45min)
Phillipe NAVEAU
Salle des conférences-IRA
Title: Extreme value analysis of records temperatures with physical constraint.
- Abstract: Heatwaves have devastating impacts on societies and ecosystems. Their frequencies and intensities are increasing globally with anthropogenic climate change. Statistical models using Extreme Value Theory (EVT) have been used for quantifying risks of extreme temperatures but recent very intense events have cast doubt on their ability to represent the tail probabilities of temperatures. Using outputs from a large ensemble of a climate model, we show that physics-based estimates of the upper-bound of temperatures in the mid-latitudes are 3--8°C higher than suggested by EVT-based models. We propose a new method to bridge the gap between the physical and statistical estimates by forcing the EVT-based models to have an upper bound coherent with the bound provided by the instability of the air column. We show that our method reduces the underestimation of tail risks while not deteriorating the
- Abstract: Heatwaves have devastating impacts on societies and ecosystems. Their frequencies and intensities are increasing globally with anthropogenic climate change. Statistical models using Extreme Value Theory (EVT) have been used for quantifying risks of extreme temperatures but recent very intense events have cast doubt on their ability to represent the tail probabilities of temperatures. Using outputs from a large ensemble of a climate model, we show that physics-based estimates of the upper-bound of temperatures in the mid-latitudes are 3--8°C higher than suggested by EVT-based models. We propose a new method to bridge the gap between the physical and statistical estimates by forcing the EVT-based models to have an upper bound coherent with the bound provided by the instability of the air column. We show that our method reduces the underestimation of tail risks while not deteriorating the
12:45 - 14:00 (1h15)
Déjeuner
14:00 - 15:15 (1h15)
Table ronde
Salle des conférences-IRA
15:15 - 15:30 (15min)
Pause café
15:30 - 16:15 (45min)
Antoine MOLL
Salle des conférences-IRA
-Title: Pricing professional athletes: a question of age? A new, more accurate and inclusive pricing method for male professional footballers based on Transfermarkt's collaborative data.
- Abstract: The football industry is generating billions in revenue, largely from matchday sales, commercial activities, and broadcast rights, player wages form a significant portion of club revenues, making their availability for games and training economically vital. Therefore, disability insurance products are essential for football clubs to mitigate the financial risks associated with player unavailability. The traditional underwriting process for professional athletes, which often relies on age and specific injury exclusions, presents coverage gaps, particularly for work stoppages due to injuries and for athletes of more than 35 years old. To address these gaps, the Transfermarkt® database has been utilized to develop a more accurate and inclusive pricing method that considers the actual
- Abstract: The football industry is generating billions in revenue, largely from matchday sales, commercial activities, and broadcast rights, player wages form a significant portion of club revenues, making their availability for games and training economically vital. Therefore, disability insurance products are essential for football clubs to mitigate the financial risks associated with player unavailability. The traditional underwriting process for professional athletes, which often relies on age and specific injury exclusions, presents coverage gaps, particularly for work stoppages due to injuries and for athletes of more than 35 years old. To address these gaps, the Transfermarkt® database has been utilized to develop a more accurate and inclusive pricing method that considers the actual
16:15 - 17:00 (45min)
Nicolas BARADEL
Salle des conférences-IRA
-Title: Continuous-time modeling and bootstrap for chain ladder reserving.
-Abstract:We revisit the famous Mack's model \cite{mack1993distribution} which gives an estimate for the mean square error of prediction of the chain ladder claims reserves. We introduce a stochastic differential equation driven by a Brownian motion to model accumulated total claims amount for the chain ladder method. Within this continuous-time framework, we propose a bootstrap technique for estimating the distribution of claims reserves. It turns out that our approach leads to inherently capturing asymmetry and non-negativity, eliminating the necessity for additional assumptions. We conclude with a case study and comparative analysis against alternative methodologies based on Mack's model.
-Abstract:We revisit the famous Mack's model \cite{mack1993distribution} which gives an estimate for the mean square error of prediction of the chain ladder claims reserves. We introduce a stochastic differential equation driven by a Brownian motion to model accumulated total claims amount for the chain ladder method. Within this continuous-time framework, we propose a bootstrap technique for estimating the distribution of claims reserves. It turns out that our approach leads to inherently capturing asymmetry and non-negativity, eliminating the necessity for additional assumptions. We conclude with a case study and comparative analysis against alternative methodologies based on Mack's model.
9:30 - 10:15 (45min)
Denys POMMERET
Salle des conférences-IRA
-Title : Estimation de copules et applications.
10:15 - 11:00 (45min)
Anne SABOURIN
Salle des conférences-IRA
-Title: On regression in extreme regions.
-Abstract: In the classic regression problem, the value of a real-valued random variable $Y$ is to be predicted based on the observation of a random vector $X$, taking its values in $\mathbb{R}^d$ with $d\geq 1$ say. The statistical learning problem consists in building a predictive function $\hat{f}:\mathbb{R}^d\to \mathbb{R}$ based on independent copies of the pair $(X,Y)$ so that $Y$ is approximated by $\hat{f}(X)$ with minimum error in the mean-squared sense. Motivated by various applications, ranging from environmental sciences to finance or insurance, special attention is paid here to the case of extreme (\textit{i.e.} very large) observations $X$. Because of their rarity, they contribute in a negligible manner to the (empirical) error and the predictive performance of empirical quadratic risk minimizers can be consequently very poor in extreme regions. In this paper, we develop a general framework for regression in the extremes.
-Abstract: In the classic regression problem, the value of a real-valued random variable $Y$ is to be predicted based on the observation of a random vector $X$, taking its values in $\mathbb{R}^d$ with $d\geq 1$ say. The statistical learning problem consists in building a predictive function $\hat{f}:\mathbb{R}^d\to \mathbb{R}$ based on independent copies of the pair $(X,Y)$ so that $Y$ is approximated by $\hat{f}(X)$ with minimum error in the mean-squared sense. Motivated by various applications, ranging from environmental sciences to finance or insurance, special attention is paid here to the case of extreme (\textit{i.e.} very large) observations $X$. Because of their rarity, they contribute in a negligible manner to the (empirical) error and the predictive performance of empirical quadratic risk minimizers can be consequently very poor in extreme regions. In this paper, we develop a general framework for regression in the extremes.
11:00 - 11:15 (15min)
Pause café
11:15 - 12:00 (45min)
Patrice BERTAIL
Salle des conférences-IRA
-Title:Biais en apprentissage statistique: une revue et une approche semi-paramétrique par calibration.
- Abstract:Dans cette présentation, nous passons d'abord en revue quelques types de biais pouvant apparaître fréquemment dans des problèmes d'apprentissage statistique : bias de données, de catégories, biais de sélection, biais d'endogénéité, biais de modèle dans les cas non-stationnaires etc... Ces types de biais sont bien connus dans la littérature statistique et plus encore en économétrie mais pas toujours pris en compte en apprentissage. Ces problèmes de biais posent non seulement des problèmes techniques mais aussi éthiques comme nous le verrons sur quelques exemples. Les techniques de corrections de biais sont souvent inspirés de technique de sondage et reposent sur des pondérations adéquates des individus : Vardi dans les année 80's a même donné des conditions nécessaires et suffisantes pour pouvoir corriger de biais (essentiellement de sélection ou d'endogénéité) en
- Abstract:Dans cette présentation, nous passons d'abord en revue quelques types de biais pouvant apparaître fréquemment dans des problèmes d'apprentissage statistique : bias de données, de catégories, biais de sélection, biais d'endogénéité, biais de modèle dans les cas non-stationnaires etc... Ces types de biais sont bien connus dans la littérature statistique et plus encore en économétrie mais pas toujours pris en compte en apprentissage. Ces problèmes de biais posent non seulement des problèmes techniques mais aussi éthiques comme nous le verrons sur quelques exemples. Les techniques de corrections de biais sont souvent inspirés de technique de sondage et reposent sur des pondérations adéquates des individus : Vardi dans les année 80's a même donné des conditions nécessaires et suffisantes pour pouvoir corriger de biais (essentiellement de sélection ou d'endogénéité) en
12:00 - 12:45 (45min)
Xavier GUEGUIN
Salle des conférences-IRA
-Title: Traitement de l’incertitude dans la prévision des résultats en assurance non vie.
12:45 - 14:00 (1h15)
Déjeuner
