About
51
Publications
10,330
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,127
Citations
Introduction
I taught at the University Paris Nanterre from 1984 to 2018. My favorite research topics are experience rating in non-life insurance and related
econometric issues. I am retired from active duty since 2019.
Current institution
Additional affiliations
July 2005 - July 2012
October 1984 - present
Université de Nanterre - Paris X
Editor roles
Education
September 1976 - June 1982
Publications
Publications (51)
A guiding theme of the chapter is the memory in longitudinal nonlife insurance data, and especially in automobile insurance. Memory in bonus–malus scales, in real-world rating structures and in statistical models on insurance data, is related to the magnitude of no-claim discounts, if the seniority is controlled for. By increasing length of memory...
A guiding theme of the chapter is the memory in longitudinal non-life insurance data, and especially in automobile insurance. Memory in bonus-malus scales, in real-world rating structures and in statistical models on insurance data is related to the magnitude of no-claim discounts, if the seniority is controlled for. By increasing length of memory...
The result given in the title and proven in this paper is of interest for the semiparametric analysis of panel data with nonnegative dependent variables, using mixture models with stationary random effects. As shown in previous articles, positivity properties of the autocovariance function of stationary random effects ensure the positivity of the a...
The following R programs illustrate several calculations reported in the paper. They are provided with comments on their inputs and outputs, and with additional mathematical formulas. P1: Mixing distributions of type ARFIMA(0, d, 0). Derivation of the partial autocorrelation coefficients with the Levinson-Durbin recursion. Derivation of the weights...
Nonnegative linear filtering of nonnegative risk variables necessitates positivity properties on the variance–covariance matrices of random effects, if experience rating is derived from mixture models. A variance–covariance matrix is a potential if it is nonsingular and if its inverse is diagonally dominant, with off-diagonal entries that are all n...
Nonnegative linear filtering of nonnegative risk variables necessitates positivity properties of the random effects, if experience rating is derived from mixture models. A variance-covariance matrix is a potential if it is nonsingular and if its inverse is diagonally dominant, with off-diagonal entries that are all nonpositive. We consider risk mod...
Allowing for the seniority of claims and of risk exposure in the prediction of frequency risks necessitates dynamic random effects in Poisson mixtures. Non-life insurance data show evidence of long memory in stationary random effects. This paper proves that the ARFIMA(0,d,0) mixtures of Poisson distributions ensure nonnegative credibilities per per...
Allowing for the seniority of claims and of risk exposure in the prediction of frequency risks necessitates dynamic random effects in Poisson mixtures. Non-life insurance data show evidence of long memory in stationary random effects. This paper proves that the ARFIMA(0, d, 0) mixtures of Poisson distributions ensure nonnegative credibilities per p...
This paper provides a toolbox for the credibility analysis of frequency risks, with allowance for the seniority of claims and of risk exposure. We use Poisson models with dynamic and second-order stationary random effects that ensure nonnegative credibilities per period. We specify classes of autocovariance functions that are compatible with positi...
Talks in Finance and Insurance, ETH Zürich
This paper provides a toolbox for the credibility analysis of frequency risks, with allowance for the seniority of claims. We use Poisson models with dynamic and positive random effects that ensure nonnegative credibilities per period. We specify classes of autocovariance functions that are compatible with such random effects. Random effects with n...
The main purpose of this study is to determine conditions under which linear credibility coefficients and the autocovariance function of random effects exhibit a similar shape for positive lags in frequency risk models. To what extent a positive and bell-shaped autocovariance function induces positive credibility coefficients is the motivation of t...
Road safety policies and automobile insurance contracts often use incentive mechanisms based on traffic violations and accidents to promote safe driving. Can these mechanisms improve road safety efficiently? Do they reduce asymmetric information between drivers and insurers and regulators? In other words, is there residual asymmetric information in...
A captive is an insurance or reinsurance company established by a parent group to finance its own risks. Captives mix internal risk pooling between the business units of the parent group and risk transfer toward the reinsurance market. We analyze captives from an optimal insurance contract perspective. The paper considers the vertical contractual c...
Facts and figures about road safety in Canada, and international comparisons
Different approaches for the cost analysis of accidents
Incentive schemes to careful driving
This paper presents statistical models which lead to experience rating in insurance. Serial correlation for risk variables can receive endogeneous or exogeneous explanations. The interpretation retained by actuarial models is exogeneous and reflects the positive contagion usually observed for the number of claims. This positive contagion can be exp...
This article presents a case study of a portfolio of individual long‐term insurance contracts sold by a Spanish mutual company. We describe the risk levels, the rating structure, and the implied cross‐subsidies on a portfolio of policies providing health, life, and long‐term care insurance. We show evidence of reclassification risk through the hist...
Conference on longevity and pensions funds Long-Term Care: risk description, lapse behavior of LTC policyholders and issu
Road safety policies often use incentive mechanisms based on traffic violations to promote safe driving-—for example, fines, experience rating, and point-record driver's licenses. We analyze the effectiveness of these mechanisms in promoting safe driving. We derive their theoretical properties with respect to contract time and accumulated demerit p...
Les travaux présentés dans cette HDR sont le reflet de mon activité de recherches effectuée au laboratoire THEMA de l'université Paris-X Nanterre de 1994 à 2005, et poursuivie depuis au département d'économie de l'Ecole Polytechnique. Ces travaux portent sur les propriétés prédictives et incitatives (à la prévention des risques) des historiques ind...
Le risque de dépendance: quelle place pour l’assurance?
Les politiques de s�curit� routi�re utilisent souvent des m�canismes incitatifs bas�s sur les infractions pour am�liorer le comportement des conducteurs. Ces m�canismes sont par exemple des amendes, des primes d'assurance ou des permis � points. Nous analysons l'efficacit� incitative de ces m�canismes. Nous obtenons leurs propri�t�s th�oriques par...
Long-Term Care (LTC): Risk Description of a Spanish Portfolio and Economic Analysis of the Timing of Insurance Purchase
Les politiques de s�curit� routi�re utilisent souvent des m�canismes incitatifs bas�s sur les infractions pour am�liorer le comportement des conducteurs. Ces m�canismes sont, soit mon�taires (amendes, primes d'assurance), soit non mon�taires (permis � points). Nous analysons l'efficacit� de ces m�canismes dans l'incitation � une conduite prudente....
This paper analyzes the rationale of long-term care insurance purchasing, from a statistical analysis of insurance data and a life cycle model. We make a short survey of the pros and cons of LTC insurance purchase. Then risk distributions in the occurrence and duration dimension are estimated on a Spanish portfolio. Calendar effects are estimated b...
This paper questions the equidistribution assumption for the random effects in a frequency risk model. Two models are presented, which use parametric and nonparametric links between the variance of the random effect and frequency risk. They are estimated on a Spanish automobile insurance portfolio, for which a decreasing link is obtained. Conclusio...
Selection bias results from a discrepancy between the range of estimation of a statistical model and its range of application. This is the case for fraud risk models, which are estimated on audited claims but applied on incoming claims in the design of auditing strategies. Now audited claims are a minority within the parent sample since they are ch...
Les politiques de sécurité routière utilisent souvent des mécanismes incitatifs basés sur les infractions pour améliorer le comportement des conducteurs. Ces mécanismes sont soit monétaires (amendes, primes d'assurance), soit non monétaires (permis à points). Nous utilisons des données québécoises couvrant une période allant de 1983 à 1996 pour ana...
Ce papier remet en question l'hypothèse d'équidistribution des effets aléatoires dans un modèle de risque fréquence. Deux modèles sont présentés, qui utilisent des liens paramétriques et non paramétriques entre la variance de l'effet aléatoire et le risque fréquence. Ils sont estimés sur un portefeuille de contrats espagnols d'assurance automobile,...
Selection bias and auditing policies on insurance claims
This paper estimates and tests autoregressive specifications for dynamic random effects in a frequency risk model. Linear credibility predictors are derived from the estimators. Examples are provided from the automobile portfolio of a Spanish insurance company.
This article proposes a computer-intensive methodology to build bonus-malus scales in automobile insurance. The claim frequency model is taken from Pinquet, Guillén, and Bolancé (2001). It accounts for overdispersion, heteroskedasticity, and dependence among repeated observations. Explanatory variables are taken into account in the determination of...
A standard problem of applied contracts theory is to empirically distinguish between adverse selection and moral hazard. We show that dynamic insurance data allow to distinguish moral hazard from dynamic selection on unobservables. In the presence of moral hazard, experience rating implies negative occurrence dependence: individual claim intensitie...
This paper exploits dynamic features of insurance contracts in the empirical analysis of moral hazard. We first show that experience rating implies negative occurrence dependence under moral hazard: individual claim intensities decrease with the number of past claims. We then show that dynamic insurance data allow to distinguish this moral-hazard e...
The purpose of the paper is to use the age of claims in the prediction of risks. A dynamic random effects model on longitudinal count data is presen- ted, and estimated on the portfolio of a major Spanish insurance company. The estimated autocorrelation coefficients of stationary random effects are decreasing. A consequence is that the predictive a...
We test the efficiency associated with the role of memory in long-term contracting. Bonus-malus schemes in automobile insurance are examples of contracts that use memory. During the eighties different contributions showed how multi-period contracting under moral hazard improves resource allocation. It was also stressed that these models did not con...
This paper presents statistical models which lead to experience rating in insurance. Serial correlation for risk variables
can receive endogeneous or exogeneous explanations. The paper recalls that the main interpretation for automobile insurance
is exogeneous, since positive contagion is always observed for the number of claims reported and since...
Abstract The purpose of the paper is to use the date of claims in the prediction of risks. Dynamic random,e¤ects models on longitudinal count data are presented, and estimated on the portfolio of a major Spanish insurance company. The estimated autocorrelation coe¢cients of stationary random e¤ects are decreasing. A consequence is that the predicti...
This paper proposes bonus-malus systems for fleets of vehicles, by using the individual characteristics of both the vehicles and the carriers. Bonus-malus coefficients are computed from the history of claims or from the history of safety offences of the carriers and the drivers. The empirical results are derived from a data set obtained from the So...
[fre] La sécurisation des prêts pour l'accession à la propriété concerne les organismes de crédit mais aussi la puissance publique, avec le développement des prêts à taux zéro. L'assurance chômage privée des prêts hypothécaires par le secteur concurrentiel est une réponse à un objectif de sécurisation, qui évite la budgétisation des prestations qu'...
Abstract This paper provides bonus-malus systems which rest on di¤erent types of claims. Consistent estimators are given for some moments of the mixing distribution of a multi equation Poisson model with random e¤ects. Bonus- malus coe¢cients are then obtained with the expected value principle, and from,linear credibility predictors. Empirical resu...
The objective of this paper is to make allowance for cost of claims in experience ra- ting. We design here a bonus-malus system for the pure premium of insurance con- tracts, from a rating based on their individual characteristics Empmcal results are presented, that are drawn from a French data base of automobde insurance contracts.
The analysis of individual risks in insurance raises problems that occur in any statistical analysis of longitudinal data. Considering insurance data, the endogeneous variables are severity variables (for instance: number and cost of claims, duration of compensations, and so on). The exogeneous variables of the current period can be first be used a...
Le but de ce papier est d'integrer, dans la tarification a posteriori de contrats d'assurance, la gravite des sinistres par la prise en compte de leur cout. Ce papier propose une methode de bonus- malus sur la prime pure des contrats, ceux-ci etant tarifes a partir de caracteristiques individuelles.
L'heterogeneite inexpliquee dans les modeles statistiques peut etre etudiee a partir d'un systeme mixte de lois conditionnelles a des composantes d'heterogeneite, et d'une famille de lois sur ces composantes. Dans le cas scalaire on calcule le score par rapport a la variance de la composante, et on montre l'invariance du lagrangien par rapport aux...
