Jérôme Dupuis

---

• Type de publi. : Pré-publication, Document de travail
• Date de publi. : 28/08/2020
• Auteurs : Jérôme Dupuis
• Organismes : Institut de Mathématiques de Toulouse UMR5219

---

Résumé : We consider the problem of estimating the size N of a closed population from q incomplete lists. Estimation of N is based on capture-recapture type models. We use graphical models to deal with possible dependencies between lists. The current parametriza-tion involves clique probabilities which have no simple concrete meaning and are delicate to manipulate in a Bayesian context insofar as hyper-Dirichlet distributions are used as priors. Our parametrization involves marginal and conditional capture probabilities. We develop our approach with q = 3. We show that there is a one-to-one and onto correspondence between both parametrizations and that placing hyper-Dirichlet distributions on the clique parameter boils down to place independent beta distributions on the capture parameters. When N is small, the non informative Bayesian analysis encounters difficulties. The posterior distribution of N may not exist for a particular graphical model: we give a necessary and sufficient condition of existence for each. Moreover, it is highly desirable that the priors on capture are compatible across the different models. Now, due to the small size of N , fulfilling this requirement demands a particular attention. We conclude by extending our approach to q = 4 lists.

---

Fichiers liés : Size-Population-Hal-2020.pdf

---

Source

---

• Type de publi. : Article dans une revue
• Date de publi. : 31/03/2019
• Auteurs : Christophe SavardAli SariPascal VenetAnni NikulinaJérôme Dupuis
• Organismes : Ampère, Département Méthodes pour l'Ingénierie des Systèmes Ampère, Département Méthodes pour l'Ingénierie des Systèmes Ampère, Département Méthodes pour l'Ingénierie des Systèmes Saint Petersburg Mining University IAE Lille University School of Management - Lille
• Publié dans Global Journal of Business, Economics and Management: Current Issues le 16/05/2019

---

Résumé : This paper links the scientific fields of electrical energy storage with that of managerial performance. It first presents the evaluation method of the efficient framework for quality management (EFQM) model, based on the criteria set interested in which way the management of an organisation acts on the results. An example fulfilling the requirements for an EFQM assessment is also summarised: the search for a high-performance battery internal architecture in order to improve its lifespan. To date, several different architectures, combined with a specific management algorithm, allow to similarly extending the battery lifespan. The method presented here helps to identify which architecture has the best management performance. In addition, it identifies its strengths and main defect.

---

Source

---

• Type de publi. : Pré-publication, Document de travail
• Date de publi. : 09/10/2017
• Auteurs : Jérôme Dupuis
• Organismes : Institut de Mathématiques de Toulouse UMR5219

---

Résumé : We consider the problem of estimating the size N of a closed population from three incomplete lists. Estimation of N is based on capture-recapture type models. Our approach uses graphical models to deal with some possible dependences between the lists. Our parametriza-tion involves marginal and conditional capture probabilities rather than clique probabilities, which facilitates the incorporation of the prior information available on capture. We link both parametriza-tions and we show that assuming an hyper-Dirichlet distribution for the clique parameter of any sub-model boils down to assume that some capture probabilities follow independently beta distributions (and conversely). Prior information on capture is incorporated via a descending procedure which guarantees that the prior distributions are, in a certain sense, compatible across the different models. As far as N is concerned, an improper prior is usually adopted for N ; as a result, the posterior distribution of N may not exist. We provide a necessary and sufficient condition for it exists, when inference is based on a Bayesian model averaging.

---

Fichiers liés : Size-Population-Octobre2017-Hal.pdf

---

Source

---

• Type de publi. : Pré-publication, Document de travail
• Date de publi. : 07/06/2017
• Auteurs : Jérôme Dupuis
• Organismes : Institut de Mathématiques de Toulouse UMR5219

---

Résumé : In Bayesian statistics, the available prior information on a statistical parameter θ is taken into account via a prior distribution. We consider the problem of incorporating prior information on a parameter θ ∈ [0, 1] consisting of a 95% credible interval and either the mean or the mode. Two prior distributions are considered for incorporated these two kinds of priors: the beta distribution and the two-sided power (TSP) distribution. It appears that the beta distribution is preferable to the TSP distribution when the prior involves the mean. The preference is reversed when the prior includes the mode. We show that, in this case, there exits a unique TSP distribution which allows us to incorporate such a prior. In all the other cases, results of existence are established, but uniqueness may not hold. That is why we advocate collecting from an expert the prior mode rather the prior mean, and incorporating the resulting prior via a TSP distribution.

---

Fichiers liés : TwoSidedPower-HalPaper-2017.pdf

---

Source

---

• Type de publi. : Chapître d'ouvrage
• Date de publi. : 26/08/2016
• Auteurs : Emil TurcMarcel GuenounRodrigues Miguel Angelo V.Demirkaya YukselJérôme Dupuis
• Organismes : Institut de management public et de gouvernance territoriale Centre d'Études et de Recherche en Gestion d'Aix-Marseille Institut de management public et de gouvernance territoriale Centre d'Études et de Recherche en Gestion d'Aix-Marseille Lille économie management - UMR 9221

---

Résumé : This chapter examines the diffusion of performance measurement and management systems (PMMS) as outposts of NPM reforms in local governments of the Napoleonic administrative cluster (France, Turkey, and Portugal). We use in-depth analyses of three municipalities selected as “paradigmatic” cases for the national contexts, meant to provide an average feel of how local governments reacted to national reforms, contexts, and variations of the Napoleonic model. Case classifications reveal little in-sample variation of the PMMS, all appearing as slight adaptations of Bouckaert and Halligan’s, (2008) “performance administration” model despite distinct national patterns of local decentralization, reforms strategies, and state mechanisms of funding, steering, and control. This conservatism may confirm the infertile ground of Napoleonic LGs for NPM values and performance reforms. Yet the variety of mechanisms at work across countries calls for attention to pending evolutions of the Napoleonic model.

---

Source

---

• Type de publi. : Pré-publication, Document de travail
• Date de publi. : 09/12/2015
• Auteurs : Jérôme DupuisJean Joachim
• Organismes : Institut de Mathématiques de Toulouse UMR5219 Unité de recherche Comportement et Ecologie de la Faune Sauvage

---

Résumé : The occupancy rate of a target species in a region divided in quadrats (or sites) is defined as the proportion of quadrats occupied by this species. This is a key quantity in site occupancy models which typically remains unknown after the data are collected, because the probability of detecting a target species in a given quadrat is < 1. Implementing tests on occupancy rates leads to a quite unusual situation, because an occupancy rate is not a statistical parameter, but a function of a discrete process partially observed. To deal with that difficulty, we adopt a Bayesian view within which the treatment of such tests turns out to be natural. We develop our approach for discrete-time site occupancy data, and we illustrate it by testing if the occupancy rate of a bird species increases over time (colonization test). A Bayesian model averaging is implemented to deal with the fact that several plausible models are viewed for the data at hand. We state a closed-form expression for the posterior probability of each model. The posterior probability of the null hypothesis (under a given model) is obtained by implementing a data augmentation algorithm. Finally, from a variety of examples, we show that the Bayesian methodology allows us to address a wide range of questions about occupancy.

---

Fichiers liés : Dupuis&Joachim-2015-Hal.pdf

---

Source

---

• Type de publi. : Article dans une revue
• Date de publi. : 01/09/2015
• Auteurs : Jérôme Dupuis
• Organismes : Lille économie management - UMR 9221
• Publié dans Revue Française de Gestion le 28/10/2020

---

Résumé : Cet article se propose d'étudier, dans le contexte caractérisant le nouveau management territorial, les rôles, positionnements et postures de managers territoriaux eu égard à trois de leurs missions exercées ; comme assistant à maître d'ouvrage, représentant du maître d'ouvrage et maître d'oeuvre, respectivement dans la conduite des politiques publiques, des projets de territoire et des organisations. En s'appuyant sur une analyse interprétative de 65 entretiens menés à deux reprises en 2012 et 2014 auprès de cadres dirigeants des collectivités locales, l'auteur montre quelles sont les tensions pouvant naître entre ces différents registres.

---

Source

---

• Type de publi. : Article dans une revue
• Date de publi. : 01/09/2015
• Auteurs : Jérôme DupuisDavid Carassus
• Organismes : Centre de recherche et d'études en gestion
• Publié dans Revue du Gestionnaire Public le 29/10/2020

---

Source

---

• Type de publi. : Chapître d'ouvrage
• Date de publi. : 01/04/2015
• Auteurs : A. CartonJérôme DupuisS. Saint-Pol
• Organismes : Lille économie management - UMR 9221

---

Source

---

• Type de publi. : Communication dans un congrès
• Date de publi. : 20/05/2014
• Auteurs : David CarassusJérôme DupuisChristophe FavoreuMarcel GuenounChristophe Maurel
• Organismes : Centre de recherche et d'études en gestion Laboratoire de Statistique et Probabilités Toulouse Business School Institut de management public et de gouvernance territoriale Centre d'Études et de Recherche en Gestion d'Aix-Marseille

---

Fichiers liés : ABS_AIRMAP_AixEnProvence_052014.pdf

---

Source