Développement d’un modèle centré sur l’individu des déplacements du caribou, du loup et de l’orignal, et de leurs interactions, en forêt boréale aménagée

Le caribou forestier est une espèce menacée au Canada, la principale hypothèse au déclin des populations étant l’intensification de la prédation provoquée par les perturbations anthropiques du paysage. Afin de faire face à cette situation, il est nécessaire d’étudier et comprendre l’impact de l’environnement sur les interactions prédateur-proies entre le caribou et le loup, ainsi qu’avec l’orignal, qui est sa principale proie alternative. Pour cela, cette thèse présente la conception d’un modèle centré sur l’individu des déplacements de ces trois espèces en fonction de leur environnement, dont résulteront les interactions prédateur-proies. Afin de permettre l’application de ce modèle sur de longues périodes, et donc pour un environnement changeant, une méthodologie a été développée, qui s’articule atour de deux aspects principaux. Tout d’abord, la notion de niveaux d’émergence est introduite, permettant d’ordonner les comportements observables du système selon leurs interdépendances, afin de choisir comme trait du modèle un com- portement correspondant au domaine d’applicabilité visé. Ordonner les comportements selon leurs niveaux d’émergence permet également d’identifier la redondance entre les patrons, qui peut être à l’origine d’un phénomène de sur-apprentissage lorsqu’ils sont utilisés lors de la calibration. Dans un second temps, un nouveau protocole pour la calibration et la validation du ou des traits choisis à l’aide des niveaux d’émergence, nommé réplication de système basé sur l’individu (Individual Based System Replication - IBSRtion) est également présenté. Ce protocole met l’emphase sur la modélisation directe, contrairement au principal protocole existant, la modélisation orientée patrons (Pattern Oriented Modelling - POM), et permet une approche empirique en générant artificiellement des données non disponibles ou ne pouvant être récoltées par des études de terrains. IBSRtion a également l’avantage de pouvoir être intégrée dans POM, afin de contribuer à la création d’une méthodologie universelle pour la conception de modèles centrés sur l’individu. Le processus de conception de ce modèle aura entre autre permis de faire une synthèse des connaissances et d’identifier certaines lacunes. Une étude visant à palier le manque de connaissances satisfaisantes sur les réponses comportementales à court-terme des proies face au risque de prédation a notamment permis d’observer que celles-ci sont une combinaison de comportements chroniques et éphémères, et que les mécanismes qui en sont à l’origine sont complexes et non-linéaires. Le résultat de ce travail est un modèle complexe utilisant de nombreux sous-modèles, et calibré de façon empirique, applicable à une grande variété d’environnements. Ce modèle a permis de tester l’impact de l’enfeuillement sur les relations prédateur-proies. Des simulations ont été effectuées pour différentes quantités d’enfeuillement, suivant deux configurations spatiales différentes. Les résultats de simulation suggèrent que des plans d’aménagement considérant également l’habitat de l’orignal pourraient être bénéfiques pour le caribou forestier, car ils permettraient d’améliorer la ségrégation spatiale entre les deux espèces, et donc entre le caribou et le loup. En le couplant avec un module de naissances et de morts naturelles ainsi qu’un modèle d’évolution du paysage, ce modèle permettra par la suite d’évaluer l’impact de plans d’aménagement forestier sur la viabilité des populations de caribou forestier.

Essays on numerically efficient inference in nonlinear and non-Gaussian state space models, and commodity market analysis.

The first two articles build procedures to simulate vector of univariate states and estimate parameters in nonlinear and non Gaussian state space models. We propose state space speci fications that offer more flexibility in modeling dynamic relationship with latent variables. Our procedures are extension of the HESSIAN method of McCausland[2012]. Thus, they use approximation of the posterior density of the vector of states that allow to : simulate directly from the state vector posterior distribution, to simulate the states vector in one bloc and jointly with the vector of parameters, and to not allow data augmentation. These properties allow to build posterior simulators with very high relative numerical efficiency. Generic, they open a new path in nonlinear and non Gaussian state space analysis with limited contribution of the modeler. The third article is an essay in commodity market analysis. Private firms coexist with farmers' cooperatives in commodity markets in subsaharan african countries. The private firms have the biggest market share while some theoretical models predict they disappearance once confronted to farmers cooperatives. Elsewhere, some empirical studies and observations link cooperative incidence in a region with interpersonal trust, and thus to farmers trust toward cooperatives. We propose a model that sustain these empirical facts. A model where the cooperative reputation is a leading factor determining the market equilibrium of a price competition between a cooperative and a private firm

Analyse du raisonnement covariationnel favorisant le passage de la fonction à la dérivée et des situations qui en sollicitent le déploiement chez des élèves de 15 à 18 ans

Afin de mieux cerner les enjeux de la transition entre le secondaire et le postsecondaire, nous proposons un examen du passage de la notion de fonction à celle de dérivée. À la lumière de plusieurs travaux mettant en évidence des difficultés inhérentes à ce passage, et nous basant sur les recherches de Carlson et ses collègues (Carlson, 2002; Carlson, Jacobs, Coe, Larsen et Hsu, 2002; Carlson, Larsen et Jacobs, 2001; Oehrtman, Carlson et Thompson, 2008) sur le raisonnement covariationnel, nous présentons une analyse de la dynamique du développement de ce raisonnement chez des petits groupes d’élèves de la fin du secondaire et du début du collégial dans quatre situations-problèmes différentes. L’analyse des raisonnements de ces groupes d’élèves nous a permis, d’une part, de raffiner la grille proposée par Carlson en mettant en évidence, non seulement des unités de processus de modélisation (ou unités de raisonnement) mises en action par ces élèves lors des activités proposées, mais aussi leurs rôles au sein de la dynamique du raisonnement. D’autre part, cette analyse révèle l’influence de certaines caractéristiques des situations sur les interactions non linéaires entre ces unités.

Dynamic Programming Approaches for Estimating and Applying Large-scale Discrete Choice Models

People go through their life making all kinds of decisions, and some of these decisions affect their demand for transportation, for example, their choices of where to live and where to work, how and when to travel and which route to take. Transport related choices are typically time dependent and characterized by large number of alternatives that can be spatially correlated. This thesis deals with models that can be used to analyze and predict discrete choices in large-scale networks. The proposed models and methods are highly relevant for, but not limited to, transport applications. We model decisions as sequences of choices within the dynamic discrete choice framework, also known as parametric Markov decision processes. Such models are known to be difficult to estimate and to apply to make predictions because dynamic programming problems need to be solved in order to compute choice probabilities. In this thesis we show that it is possible to explore the network structure and the flexibility of dynamic programming so that the dynamic discrete choice modeling approach is not only useful to model time dependent choices, but also makes it easier to model large-scale static choices. The thesis consists of seven articles containing a number of models and methods for estimating, applying and testing large-scale discrete choice models. In the following we group the contributions under three themes: route choice modeling, large-scale multivariate extreme value (MEV) model estimation and nonlinear optimization algorithms. Five articles are related to route choice modeling. We propose different dynamic discrete choice models that allow paths to be correlated based on the MEV and mixed logit models. The resulting route choice models become expensive to estimate and we deal with this challenge by proposing innovative methods that allow to reduce the estimation cost. For example, we propose a decomposition method that not only opens up for possibility of mixing, but also speeds up the estimation for simple logit models, which has implications also for traffic simulation. Moreover, we compare the utility maximization and regret minimization decision rules, and we propose a misspecification test for logit-based route choice models. The second theme is related to the estimation of static discrete choice models with large choice sets. We establish that a class of MEV models can be reformulated as dynamic discrete choice models on the networks of correlation structures. These dynamic models can then be estimated quickly using dynamic programming techniques and an efficient nonlinear optimization algorithm. Finally, the third theme focuses on structured quasi-Newton techniques for estimating discrete choice models by maximum likelihood. We examine and adapt switching methods that can be easily integrated into usual optimization algorithms (line search and trust region) to accelerate the estimation process. The proposed dynamic discrete choice models and estimation methods can be used in various discrete choice applications. In the area of big data analytics, models that can deal with large choice sets and sequential choices are important. Our research can therefore be of interest in various demand analysis applications (predictive analytics) or can be integrated with optimization models (prescriptive analytics). Furthermore, our studies indicate the potential of dynamic programming techniques in this context, even for static models, which opens up a variety of future research directions.

Physical and chemical kinetics of structural build-up of cement suspensions

Abstract : The structural build-up of fresh cement-based materials has a great impact on their structural performance after casting. Accordingly, the mixture design should be tailored to adapt the kinetics of build-up given the application on hand. The rate of structural build-up of cement-based suspensions at rest is a complex phenomenon affected by both physical and chemical structuration processes. The structuration kinetics are strongly dependent on the mixture’s composition, testing parameters, as well as the shear history. Accurate measurements of build-up rely on the efficiency of the applied pre-shear regime to achieve an initial well-dispersed state as well as the applied stress during the liquid-solid transition. Studying the physical and chemical mechanisms of build-up of cement suspensions at rest can enhance the fundamental understanding of this phenomenon. This can, therefore, allow a better control of the rheological and time-dependent properties of cement-based materials. The research focused on the use of dynamic rheology in investigating the kinetics of structural build-up of fresh cement pastes. The research program was conducted in three different phases. The first phase was devoted to evaluating the dispersing efficiency of various disruptive shear techniques. The investigated shearing profiles included rotational, oscillatory, and combination of both. The initial and final states of suspension’s structure, before and after disruption, were determined by applying a small-amplitude oscillatory shear (SAOS). The difference between the viscoelastic values before and after disruption was used to express the degree of dispersion. An efficient technique to disperse concentrated cement suspensions was developed. The second phase aimed to establish a rheometric approach to dissociate and monitor the individual physical and chemical mechanisms of build-up of cement paste. In this regard, the non-destructive dynamic rheometry was used to investigate the evolutions of both storage modulus and phase angle of inert calcium carbonate and cement suspensions. Two independent build-up indices were proposed. The structural build-up of various cement suspensions made with different cement contents, silica fume replacement percentages, and high-range water reducer dosages was evaluated using the proposed indices. These indices were then compared to the well-known thixotropic index (Athix.). Furthermore, the proposed indices were correlated to the decay in lateral pressure determined for various cement pastes cast in a pressure column. The proposed pre-shearing protocol and build-up indices (phases 1 and 2) were then used to investigate the effect of mixture’s parameters on the kinetics of structural build-up in phase 3. The investigated mixture’s parameters included cement content and fineness, alkali sulfate content, and temperature of cement suspension. Zeta potential, calorimetric, spectrometric measurements were performed to explore the corresponding microstructural changes in cement suspensions, such as inter-particle cohesion, rate of Brownian flocculation, and nucleation rate. A model linking the build-up indices and the microstructural characteristics was developed to predict the build-up behaviour of cement-based suspensions The obtained results showed that oscillatory shear may have a greater effect on dispersing concentrated cement suspension than the rotational shear. Furthermore, the increase in induced shear strain was found to enhance the breakdown of suspension’s structure until a critical point, after which thickening effects dominate. An effective dispersing method is then proposed. This consists of applying a rotational shear around the transitional value between the linear and non-linear variations of the apparent viscosity with shear rate, followed by an oscillatory shear at the crossover shear strain and high angular frequency of 100 rad/s. Investigating the evolutions of viscoelastic properties of inert calcite-based and cement suspensions and allowed establishing two independent build-up indices. The first one (the percolation time) can represent the rest time needed to form the elastic network. On the other hand, the second one (rigidification rate) can describe the increase in stress-bearing capacity of formed network due to cement hydration. In addition, results showed that combining the percolation time and the rigidification rate can provide deeper insight into the structuration process of cement suspensions. Furthermore, these indices were found to be well-correlated to the decay in the lateral pressure of cement suspensions. The variations of proposed build-up indices with mixture’s parameters showed that the percolation time is most likely controlled by the frequency of Brownian collisions, distance between dispersed particles, and intensity of cohesion between cement particles. On the other hand, a higher rigidification rate can be secured by increasing the number of contact points per unit volume of paste, nucleation rate of cement hydrates, and intensity of inter-particle cohesion.

Dynamic Programming Approaches for Estimating and Applying Large-scale Discrete Choice Models

People go through their life making all kinds of decisions, and some of these decisions affect their demand for transportation, for example, their choices of where to live and where to work, how and when to travel and which route to take. Transport related choices are typically time dependent and characterized by large number of alternatives that can be spatially correlated. This thesis deals with models that can be used to analyze and predict discrete choices in large-scale networks. The proposed models and methods are highly relevant for, but not limited to, transport applications. We model decisions as sequences of choices within the dynamic discrete choice framework, also known as parametric Markov decision processes. Such models are known to be difficult to estimate and to apply to make predictions because dynamic programming problems need to be solved in order to compute choice probabilities. In this thesis we show that it is possible to explore the network structure and the flexibility of dynamic programming so that the dynamic discrete choice modeling approach is not only useful to model time dependent choices, but also makes it easier to model large-scale static choices. The thesis consists of seven articles containing a number of models and methods for estimating, applying and testing large-scale discrete choice models. In the following we group the contributions under three themes: route choice modeling, large-scale multivariate extreme value (MEV) model estimation and nonlinear optimization algorithms. Five articles are related to route choice modeling. We propose different dynamic discrete choice models that allow paths to be correlated based on the MEV and mixed logit models. The resulting route choice models become expensive to estimate and we deal with this challenge by proposing innovative methods that allow to reduce the estimation cost. For example, we propose a decomposition method that not only opens up for possibility of mixing, but also speeds up the estimation for simple logit models, which has implications also for traffic simulation. Moreover, we compare the utility maximization and regret minimization decision rules, and we propose a misspecification test for logit-based route choice models. The second theme is related to the estimation of static discrete choice models with large choice sets. We establish that a class of MEV models can be reformulated as dynamic discrete choice models on the networks of correlation structures. These dynamic models can then be estimated quickly using dynamic programming techniques and an efficient nonlinear optimization algorithm. Finally, the third theme focuses on structured quasi-Newton techniques for estimating discrete choice models by maximum likelihood. We examine and adapt switching methods that can be easily integrated into usual optimization algorithms (line search and trust region) to accelerate the estimation process. The proposed dynamic discrete choice models and estimation methods can be used in various discrete choice applications. In the area of big data analytics, models that can deal with large choice sets and sequential choices are important. Our research can therefore be of interest in various demand analysis applications (predictive analytics) or can be integrated with optimization models (prescriptive analytics). Furthermore, our studies indicate the potential of dynamic programming techniques in this context, even for static models, which opens up a variety of future research directions.

No self-interaction for two-column massless fields

We investigate the problem of introducing consistent self-couplings in free theories for mixed tensor gauge fields whose symmetry properties are characterized by Young diagrams made of two columns of arbitrary (but different) lengths. We prove that, in flat space, these theories admit no local, Poincaré-invariant, smooth, selfinteracting deformation with at most two derivatives in the Lagrangian. Relaxing the derivative and Lorentz-invariance assumptions, there still is no deformation that modifies the gauge algebra, and in most cases no deformation that alters the gauge transformations. Our approach is based on a Becchi-Rouet-Stora-iyutin (BRST) -cohomology deformation procedure. © 2005 American Institute of Physics.

On the problems of object restoration and image extrapolation in optics

In this paper we consider the problems of object restoration and image extrapolation, according to the regularization theory of improperly posed problems. In order to take into account the stochastic nature of the noise and to introduce the main concepts of information theory, great attention is devoted to the probabilistic methods of regularization. The kind of the restored continuity is investigated in detail; in particular we prove that, while the image extrapolation presents a Hölder type stability, the object restoration has only a logarithmic continuity. © 1979 American Institute of Physics.

Supercontinuum generation in air-silica microstructured fibers with nanosecond and femtosecond pulse pumping

We study the generation of supercontinua in air-silica microstructured fibers by both nanosecond and femtosecond pulse excitation. In the nanosecond experiments, a 300-nm broadband visible continuum was generated in a 1.8-m length of fiber pumped at 532 nm by 0.8-ns pulses from a frequency-doubled passively Q-switched Nd:YAG microchip laser. At this wavelength, the dominant mode excited under the conditions of continuum generation is the LP 11 mode, and, with nanosecond pumping, self-phase modulation is negligible and the continuum generation is dominated by the interplay of Raman and parametric effects. The spectral extent of the continuum is well explained by calculations of the parametric gain curves for four-wave mixing about the zero-dispersion wavelength of the LP11 mode. In the femtosecond experiments, an 800-nm broad-band visible and near-infrared continuum has been generated in a 1-m length of fiber pumped at 780 nm by 100-fs pulses from a Kerr-lens model-locked Ti:sapphire laser. At this wavelength, excitation and continuum generation occur in the LP01 mode, and the spectral width of the observed continuum is shown to be consistent with the phase-matching bandwidth for parametric processes calculated for this fiber mode. In addition, numerical simulations based on an extended nonlinear Schrödinger equation were used to model supercontinuum generation in the femtosecond regime, with the simulation results reproducing the major features of the experimentally observed spectrum. © 2002 Optical Society of America.

Simulation-Based Finite and Large Sample Tests in Multivariate Regressions.

In the context of multivariate linear regression (MLR) models, it is well known that commonly employed asymptotic test criteria are seriously biased towards overrejection. In this paper, we propose a general method for constructing exact tests of possibly nonlinear hypotheses on the coefficients of MLR systems. For the case of uniform linear hypotheses, we present exact distributional invariance results concerning several standard test criteria. These include Wilks' likelihood ratio (LR) criterion as well as trace and maximum root criteria. The normality assumption is not necessary for most of the results to hold. Implications for inference are two-fold. First, invariance to nuisance parameters entails that the technique of Monte Carlo tests can be applied on all these statistics to obtain exact tests of uniform linear hypotheses. Second, the invariance property of the latter statistic is exploited to derive general nuisance-parameter-free bounds on the distribution of the LR statistic for arbitrary hypotheses. Even though it may be difficult to compute these bounds analytically, they can easily be simulated, hence yielding exact bounds Monte Carlo tests. Illustrative simulation experiments show that the bounds are sufficiently tight to provide conclusive results with a high probability. Our findings illustrate the value of the bounds as a tool to be used in conjunction with more traditional simulation-based test methods (e.g., the parametric bootstrap) which may be applied when the bounds are not conclusive.

Credibility and Signaling in Disinflation- a Cross Country Examination

This paper develops a model of money demand where the opportunity cost of holding money is subject to regime changes. The regimes are fully characterized by the mean and variance of inflation and are assumed to be the result of alternative government policies. Agents are unable to directly observe whether government actions are indeed consistent with the inflation rate targeted as part of a stabilization program but can construct probability inferences on the basis of available observations of inflation and money growth. Government announcements are assumed to provide agents with additional, possibly truthful information regarding the regime. This specification is estimated and tested using data from the Israeli and Argentine high inflation periods. Results indicate the successful stabilization program implemented in Israel in July 1985 was more credible than either the earlier Israeli attempt in November 1984 or the Argentine programs. Government’s signaling might substantially simplify the inference problem and increase the speed of learning on the part of the agents. However, under certain conditions, it might increase the volatility of inflation. After the introduction of an inflation stabilization plan, the welfare gains from a temporary increase in real balances might be high enough to induce agents to raise their real balances in the short-term, even if they are uncertain about the nature of government policy and the eventual outcome of the stabilization attempt. Statistically, the model restrictions cannot be rejected at the 1% significance level.

Finite-Sample Inference Methods for Simultaneous Equations and Models with Unobserved and Generated Regressors

We propose finite sample tests and confidence sets for models with unobserved and generated regressors as well as various models estimated by instrumental variables methods. The validity of the procedures is unaffected by the presence of identification problems or \"weak instruments\", so no detection of such problems is required. We study two distinct approaches for various models considered by Pagan (1984). The first one is an instrument substitution method which generalizes an approach proposed by Anderson and Rubin (1949) and Fuller (1987) for different (although related) problems, while the second one is based on splitting the sample. The instrument substitution method uses the instruments directly, instead of generated regressors, in order to test hypotheses about the \"structural parameters\" of interest and build confidence sets. The second approach relies on \"generated regressors\", which allows a gain in degrees of freedom, and a sample split technique. For inference about general possibly nonlinear transformations of model parameters, projection techniques are proposed. A distributional theory is obtained under the assumptions of Gaussian errors and strictly exogenous regressors. We show that the various tests and confidence sets proposed are (locally) \"asymptotically valid\" under much weaker assumptions. The properties of the tests proposed are examined in simulation experiments. In general, they outperform the usual asymptotic inference methods in terms of both reliability and power. Finally, the techniques suggested are applied to a model of Tobin’s q and to a model of academic performance.

Explaining the Transition Between Exchange Rate Regimes

This paper studies the transition between exchange rate regimes using a Markov chain model with time-varying transition probabilities. The probabilities are parameterized as nonlinear functions of variables suggested by the currency crisis and optimal currency area literature. Results using annual data indicate that inflation, and to a lesser extent, output growth and trade openness help explain the exchange rate regime transition dynamics.

Nonlinear Monetary Policy Rules: Some New Evidence for the U.S.

This paper derives optimal monetary policy rules in setups where certainty equivalence does not hold because either central bank preferences are not quadratic, and/or the aggregate supply relation is nonlinear. Analytical results show that these features lead to sign and size asymmetries, and nonlinearities in the policy rule. Reduced-form estimates indicate that US monetary policy can be characterized by a nonlinear policy rule after 1983, but not before 1979. This finding is consistent with the view that the Fed's inflation preferences during the Volcker-Greenspan regime differ considerably from the ones during the Burns-Miller regime.

Théorèmes d'existence pour des systèmes d'équations différentielles et d'équations aux échelles de temps.

Nous présentons dans cette thèse des théorèmes d’existence pour des systèmes d’équations différentielles non-linéaires d’ordre trois, pour des systèmes d’équa- tions et d’inclusions aux échelles de temps non-linéaires d’ordre un et pour des systèmes d’équations aux échelles de temps non-linéaires d’ordre deux sous cer- taines conditions aux limites. Dans le chapitre trois, nous introduirons une notion de tube-solution pour obtenir des théorèmes d’existence pour des systèmes d’équations différentielles du troisième ordre. Cette nouvelle notion généralise aux systèmes les notions de sous- et sur-solutions pour le problème aux limites de l’équation différentielle du troisième ordre étudiée dans [34]. Dans la dernière section de ce chapitre, nous traitons les systèmes d’ordre trois lorsque f est soumise à une condition de crois- sance de type Wintner-Nagumo. Pour admettre l’existence de solutions d’un tel système, nous aurons recours à la théorie des inclusions différentielles. Ce résultat d’existence généralise de diverses façons un théorème de Grossinho et Minhós [34]. Le chapitre suivant porte sur l’existence de solutions pour deux types de sys- tèmes d’équations aux échelles de temps du premier ordre. Les résultats d’exis- tence pour ces deux problèmes ont été obtenus grâce à des notions de tube-solution adaptées à ces systèmes. Le premier théorème généralise entre autre aux systèmes et à une échelle de temps quelconque, un résultat obtenu pour des équations aux différences finies par Mawhin et Bereanu [9]. Ce résultat permet également d’obte- nir l’existence de solutions pour de nouveaux systèmes dont on ne pouvait obtenir l’existence en utilisant le résultat de Dai et Tisdell [17]. Le deuxième théorème de ce chapitre généralise quant à lui, sous certaines conditions, des résultats de [60]. Le chapitre cinq aborde un nouveau théorème d’existence pour un système d’in- clusions aux échelles de temps du premier ordre. Selon nos recherches, aucun résultat avant celui-ci ne traitait de l’existence de solutions pour des systèmes d’inclusions de ce type. Ainsi, ce chapitre ouvre de nouvelles possibilités dans le domaine des inclusions aux échelles de temps. Notre résultat a été obtenu encore une fois à l’aide d’une hypothèse de tube-solution adaptée au problème. Au chapitre six, nous traitons l’existence de solutions pour des systèmes d’équations aux échelles de temps d’ordre deux. Le premier théorème d’existence que nous obtenons généralise les résultats de [36] étant donné que l’hypothèse que ces auteurs utilisent pour faire la majoration a priori est un cas particulier de notre hypothèse de tube-solution pour ce type de systèmes. Notons également que notre définition de tube-solution généralise aux systèmes les notions de sous- et sur-solutions introduites pour les équations d’ordre deux par [4] et [55]. Ainsi, nous généralisons également des résultats obtenus pour des équations aux échelles de temps d’ordre deux. Finalement, nous proposons un nouveau résultat d’exis- tence pour un système dont le membre droit des équations dépend de la ∆-dérivée de la fonction.