Jean-pierre Florens - Academia.edu (original) (raw)
Uploads
Papers by Jean-pierre Florens
In this paper we propose a novel method to construct confidence intervals in a class of linear in... more In this paper we propose a novel method to construct confidence intervals in a class of linear inverse problems. First, point estimators are obtained via a spectral cut-off method depending on a regularisation parameter α, that determines the bias of the estimator. Next, the proposed confidence interval corrects for this bias by explicitly estimating it based on a second regularisation parameter ρ, which is asymptotically smaller than α. The coverage error of the interval is shown to converge to zero. The proposed method is illustrated via two simulation studies, one in the context of func- tional linear regression, and the second one in the context of instrumental regression
The objective of the paper is to draw the theory of endogeneity in dynamic models in discrete and... more The objective of the paper is to draw the theory of endogeneity in dynamic models in discrete and con-tinuous time, in particular for diffusions and counting processes. We first provide an extension of the separable set-up to a separable dynamic framework given in term of semi-martingale decompositions. Then we estimate our function of interest as a stopping time for an additional noise process, whose role is played by a Brownian motion for diffusions, and a Poisson process for counting processes.
E. Vytlacil for useful discussions on this topic. The reduction of this paper had bene¯ted of num... more E. Vytlacil for useful discussions on this topic. The reduction of this paper had bene¯ted of numerous remarks and comments of L.P. Hansen 2University of Toulouse (IDEI-GREMAQ), Manufacture des Tabacs, 21, all¶ee
The nonparametric estimation of a regression function from conditional moment restrictions involv... more The nonparametric estimation of a regression function from conditional moment restrictions involving instrumental variables is considered. The rate of convergence of penalized estimators is studied in the case where the regression function is not identified from the conditional moment restriction. We also study the gain of modifying the penalty in the estimation, considering derivatives in penalty. We analyze the effect of this modification on the identification of the regression function and the rate of convergence of its estimator.
All in-text references underlined in blue are linked to publications on ResearchGate, letting you... more All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.
The problem of identification and partial identification in econometrics is considered. We carry ... more The problem of identification and partial identification in econometrics is considered. We carry out a comprehensive analysis of the identification issue from both a classical and a Bayesian point of view. We review three concepts of identification: sampling, measurable and Bayesian identification, and the relationship existing among them. We analyze the concept of exact estimability and its link with Bayesian identification. Examples are provided where measurable identification fails while Bayesian identification does not. Many classical examples of the partial identification literature are reconsidered by using a nonparametric Bayesian approach.
Point processes are a mathematical formalisation which allows one to describe individual mobiliti... more Point processes are a mathematical formalisation which allows one to describe individual mobilities or transitions between a finite number of states through time. They are particularly useful for the micro-econometric analysis of labour market dynamics. Labour statistics are often concerned with samples of workers' histories from which the econometrician can retrospectively observe individual transitions between distinct participation states: employment in a stable job (i. e., with an unlimited duration labour contract), employment in an unstable job (i. e., with a limited duration contract), unemployment (with or without eligibility to unemployment insurance systems), education, training, out-of-labour force, etc. 1 More generally, the statistical analysis of life histories data (including information on the evolution of individual matrimonial status, number of children, place of residence, etc.) is based on the theory of point processes. Besides econometrics, biostatistics, re...
Journal of Business & Economic Statistics, 2019
Sociologie du travail, 1990
Les auteurs ont analysé, sur des «cohortes de chômeurs », les passages entre différentes position... more Les auteurs ont analysé, sur des «cohortes de chômeurs », les passages entre différentes positions possibles sur le marché du travail et la durée passée dans chacune des positions. Economètres, ils ont utilisé leurs méthodes, et nous livrent ici les résultats de leurs calculs et de leurs modélisations. Ces résultats sont souvent nouveaux, comme par exemple le lien entre la durée (et le niveau) de l'indemnisation du chômage et la durée du chômage avant reprise d'un emploi ou avant radiation. Ils sont aussi parfois contre-intuitifs : ils diffèrent sensiblement de ceux qui pourraient être tirés d'une lecture des statistiques ponctuelles existantes.
Models frequently used for car insurance portfolios assume that the distribution of the expcted n... more Models frequently used for car insurance portfolios assume that the distribution of the expcted number of accident caused by a client is a mixture of Dirichlet Processes. Given that the number of clients is typically rather large, insurance companies may find it relevant to inquire wheter the posterior distribution (given the number of individual accidents) of the mixture, of the mixing parameters and the predictive distribution of a new observation, given the number of individual accidents, are consistently convergent. For most models of this rather general class, answering these questions requires unfathomable manipulations. In this paper, it is shown that answers are "easily" obtainable from general results, presented in the authors' monograph Elements of Bayesian Statistics. these answers rely on no specific assumption on the form of the distribution but only on simple conditions on latent process. Some problems of approximations are also discussed.
This paper presents recent econometric models of individual mobility on the labor market. The gen... more This paper presents recent econometric models of individual mobility on the labor market. The general framework is given by the theory of point processes which are continuous time finite state space processes. The first part of the paper is devoted to specification problems and the inference of the introduced models is not treated. The last part of the paper consider the introduction of unobservable heterogeneity and of endogeneous sampling in specification and inference analysis.
Économie & prévision, 1998
[fre] Équilibre approximatif et règle intuitive : une application aux appels d&#3... more [fre] Équilibre approximatif et règle intuitive : une application aux appels d'offres dans l'industrie spatiale par Olivier Armantier, Jean-Pierre Florens et Jean-François Richard . Nous adoptons dans cet article la notion d'équilibre approximatif obtenu à partir d'une "règle intuitive". Cette notion repose sur l'idée qu'à la place des stratégies complexes à l'équilibre de Nash, les agents utilisent des règles de
Advanced Studies in Theoretical and Applied Econometrics
Postal and Delivery Services, 2002
Econometric Modeling and Inference
This work has been prepared as chapter 19 for the volume Econometrics of Panel Data (2nd ed., Mat... more This work has been prepared as chapter 19 for the volume Econometrics of Panel Data (2nd ed., Matyas and Sevestre eds., Kluwer Academic Publishers). The main emphasis is on modelling whereas inference problems are only sketched. After a short review on the basic issues on marginal models (i.e. models involving a duration variable only) conditional models are reviewed, namely the proportional hazard model and the accelerated time model. Problems of agregation and of heterogeneity are next examined. Competing risks models are presented with a particular emphasis on the identification of dependence among latent durations. Some examples of dependent durations are given. Finally, right-censoring is treated in the framework of competing riks models.
In this paper we propose a novel method to construct confidence intervals in a class of linear in... more In this paper we propose a novel method to construct confidence intervals in a class of linear inverse problems. First, point estimators are obtained via a spectral cut-off method depending on a regularisation parameter α, that determines the bias of the estimator. Next, the proposed confidence interval corrects for this bias by explicitly estimating it based on a second regularisation parameter ρ, which is asymptotically smaller than α. The coverage error of the interval is shown to converge to zero. The proposed method is illustrated via two simulation studies, one in the context of func- tional linear regression, and the second one in the context of instrumental regression
The objective of the paper is to draw the theory of endogeneity in dynamic models in discrete and... more The objective of the paper is to draw the theory of endogeneity in dynamic models in discrete and con-tinuous time, in particular for diffusions and counting processes. We first provide an extension of the separable set-up to a separable dynamic framework given in term of semi-martingale decompositions. Then we estimate our function of interest as a stopping time for an additional noise process, whose role is played by a Brownian motion for diffusions, and a Poisson process for counting processes.
E. Vytlacil for useful discussions on this topic. The reduction of this paper had bene¯ted of num... more E. Vytlacil for useful discussions on this topic. The reduction of this paper had bene¯ted of numerous remarks and comments of L.P. Hansen 2University of Toulouse (IDEI-GREMAQ), Manufacture des Tabacs, 21, all¶ee
The nonparametric estimation of a regression function from conditional moment restrictions involv... more The nonparametric estimation of a regression function from conditional moment restrictions involving instrumental variables is considered. The rate of convergence of penalized estimators is studied in the case where the regression function is not identified from the conditional moment restriction. We also study the gain of modifying the penalty in the estimation, considering derivatives in penalty. We analyze the effect of this modification on the identification of the regression function and the rate of convergence of its estimator.
All in-text references underlined in blue are linked to publications on ResearchGate, letting you... more All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.
The problem of identification and partial identification in econometrics is considered. We carry ... more The problem of identification and partial identification in econometrics is considered. We carry out a comprehensive analysis of the identification issue from both a classical and a Bayesian point of view. We review three concepts of identification: sampling, measurable and Bayesian identification, and the relationship existing among them. We analyze the concept of exact estimability and its link with Bayesian identification. Examples are provided where measurable identification fails while Bayesian identification does not. Many classical examples of the partial identification literature are reconsidered by using a nonparametric Bayesian approach.
Point processes are a mathematical formalisation which allows one to describe individual mobiliti... more Point processes are a mathematical formalisation which allows one to describe individual mobilities or transitions between a finite number of states through time. They are particularly useful for the micro-econometric analysis of labour market dynamics. Labour statistics are often concerned with samples of workers' histories from which the econometrician can retrospectively observe individual transitions between distinct participation states: employment in a stable job (i. e., with an unlimited duration labour contract), employment in an unstable job (i. e., with a limited duration contract), unemployment (with or without eligibility to unemployment insurance systems), education, training, out-of-labour force, etc. 1 More generally, the statistical analysis of life histories data (including information on the evolution of individual matrimonial status, number of children, place of residence, etc.) is based on the theory of point processes. Besides econometrics, biostatistics, re...
Journal of Business & Economic Statistics, 2019
Sociologie du travail, 1990
Les auteurs ont analysé, sur des «cohortes de chômeurs », les passages entre différentes position... more Les auteurs ont analysé, sur des «cohortes de chômeurs », les passages entre différentes positions possibles sur le marché du travail et la durée passée dans chacune des positions. Economètres, ils ont utilisé leurs méthodes, et nous livrent ici les résultats de leurs calculs et de leurs modélisations. Ces résultats sont souvent nouveaux, comme par exemple le lien entre la durée (et le niveau) de l'indemnisation du chômage et la durée du chômage avant reprise d'un emploi ou avant radiation. Ils sont aussi parfois contre-intuitifs : ils diffèrent sensiblement de ceux qui pourraient être tirés d'une lecture des statistiques ponctuelles existantes.
Models frequently used for car insurance portfolios assume that the distribution of the expcted n... more Models frequently used for car insurance portfolios assume that the distribution of the expcted number of accident caused by a client is a mixture of Dirichlet Processes. Given that the number of clients is typically rather large, insurance companies may find it relevant to inquire wheter the posterior distribution (given the number of individual accidents) of the mixture, of the mixing parameters and the predictive distribution of a new observation, given the number of individual accidents, are consistently convergent. For most models of this rather general class, answering these questions requires unfathomable manipulations. In this paper, it is shown that answers are "easily" obtainable from general results, presented in the authors' monograph Elements of Bayesian Statistics. these answers rely on no specific assumption on the form of the distribution but only on simple conditions on latent process. Some problems of approximations are also discussed.
This paper presents recent econometric models of individual mobility on the labor market. The gen... more This paper presents recent econometric models of individual mobility on the labor market. The general framework is given by the theory of point processes which are continuous time finite state space processes. The first part of the paper is devoted to specification problems and the inference of the introduced models is not treated. The last part of the paper consider the introduction of unobservable heterogeneity and of endogeneous sampling in specification and inference analysis.
Économie & prévision, 1998
[fre] Équilibre approximatif et règle intuitive : une application aux appels d&#3... more [fre] Équilibre approximatif et règle intuitive : une application aux appels d'offres dans l'industrie spatiale par Olivier Armantier, Jean-Pierre Florens et Jean-François Richard . Nous adoptons dans cet article la notion d'équilibre approximatif obtenu à partir d'une "règle intuitive". Cette notion repose sur l'idée qu'à la place des stratégies complexes à l'équilibre de Nash, les agents utilisent des règles de
Advanced Studies in Theoretical and Applied Econometrics
Postal and Delivery Services, 2002
Econometric Modeling and Inference
This work has been prepared as chapter 19 for the volume Econometrics of Panel Data (2nd ed., Mat... more This work has been prepared as chapter 19 for the volume Econometrics of Panel Data (2nd ed., Matyas and Sevestre eds., Kluwer Academic Publishers). The main emphasis is on modelling whereas inference problems are only sketched. After a short review on the basic issues on marginal models (i.e. models involving a duration variable only) conditional models are reviewed, namely the proportional hazard model and the accelerated time model. Problems of agregation and of heterogeneity are next examined. Competing risks models are presented with a particular emphasis on the identification of dependence among latent durations. Some examples of dependent durations are given. Finally, right-censoring is treated in the framework of competing riks models.