No description available.

In this thesis, first we study the unique classical solution of quasi-linear second order parabolic partial differential equations (PDEs). For this, we study the existence and uniqueness of the $L^2_{\rho}( \mathbb{R}^{d}; \mathbb{R}^{d}) \otimes L^2_{\rho}( \mathbb{R}^{d}; \mathbb{R}^{k})\otimes L^2_{\rho}( \mathbb{R}^{d}; \mathbb{R}^{k\times d})$ valued solution of forward backward stochastic differential equations (FBSDEs) with finite horizon, the regularity property of the solution of FBSDEs and the connection between the solution of FBSDEs and the solution of quasi-linear parabolic PDEs. Then we establish their connection in the Sobolev weak sense, in order to give the weak solution of the quasi-linear parabolic PDEs. Finally, we study the unique weak solution of quasi-linear second order elliptic PDEs through the stationary solution of the FBSDEs with infinite horizon.

Na literatura de controle, os controladores preditivos disponíveis apresentam aplicações limitadas em processos com características integradoras. Uma dessas limitações está relacionada às propriedades estabilizantes que surgem da lei de controle quando o sistema é colocado em malha fechada. O principal objetivo dessa dissertação é desenvolver um algoritmo de controle preditivo, nominalmente estável, baseado num horizonte de predição infinito das saídas do sistema. Para essa finalidade, uma representação existente em variáveis de estado é analisada e estendida para ser capaz de ser aplicável para processos estáveis e integradores, com tempo morto. Dois métodos distintos que garantem a estabilidade de sistemas integradores foram desenvolvidos com um MPC de horizonte infinito. Para o caso de realimentação de estado, a estabilidade nominal do MPC proposto foi provada para os dois métodos e a eficiência dos controladores foi verificada através de exemplos da literatura de controle. Um dos controladores foi também aplicado em um processo de destilação de uma refinaria de petróleo. O algoritmo desenvolvido foi estendido para o caso de realimentação das saídas com a inclusão do filtro de Kalman para a estimativa dos estados do modelo. A aplicação do controlador em um sistema com diferentes tempos mortos entre as variáveis integradoras foi também estudada. / In the control literature, the available predictive controllers show limitations to be applied in processes that have integrating behavior. One of these limitations is related to the stabilizing properties of the resulting control law when the system is in closed loop. The main objective of this work is to develop a nominal stable control algorithm based on infinite prediction of system outputs. For this purpose an existing state space representation is explored and extended in order to be able to be applied to stable and integrating process with dead time. Two distinct methods to assure stability of integrating systems were developed for the infinite horizon MPC. For the case of state feedback, nominal stability of the resulting MPC was proved for the two methods and the efficiency of the controllers was verified through examples of the control literature. One of the controllers was also applied to an industrial distillation process of a petroleum refinery. The developed algorithm was extended to the output feedback operation with the inclusion of a Kalman filter for estimation of the model states. The application of the controller to a system with different dead times between integrating variables was also studied.

This thesis is divided into two parts. Part I deals with demand management. For goods with no substitutes, under supply constraints, fairness considerations introduce negative externalities and lead to a market failure. One example of such a good with no substitutes is water. In case of a market failure, it is necessary to design coordination mechanisms called contracts which provide the right incentives for coordination. As “repetition can yield coordination”, the aim in this part is to design price based dynamic demand management contracts which, under supply constraints, mitigate the market failure. In these contracts, we consider complete information settings; and use the status quo proposition as a fairness criterion for designing them. The contracts are designed as almost noncooperative dynamic games, within the agency theory framework, where the agent (the consumer) is induced to consume at a specified consumption level based on the incentive mechanism offered by the principal (the producer). These contracts use the solution concept of sub-game perfect Nash equilibrium (SPNE) to compute the price (mal-incentive) that acts as a credible threat for deviation from the specified consumption level. In these contracts, unlike the dynamic contracts with asymmetric information, the penalty for deviation is proportional to the amount of deviation. First, we consider a two-period demand management contract for a single consumer satisfying the status quo proposition. Under the assumption that the gain to the consumer and the loss to the producer by deviation is small, the contract is shown to be economically efficient. It is shown that, in the finite horizon, a fair demand management contract cannot be efficient. The demand management contract is homeomorphic to finite horizon alternating bargaining model. In the finite horizon alternating bargaining model, there is a unique SPNE, in which the player who offers last is always at an advantageous position. In the two-period contract, the assumption considered attenuates the last mover advantage and leads to the efficiency. We have shown that one possible way to achieve efficiency, without the assumption, is to make the agents uncertain about the period of interaction. This possibility can be included in an infinite horizon contract. Hence, next, we design an infinite horizon contract for a single consumer. It is proved that this contract is economically efficient and provides revenue sufficiency. The sensitivity analysis of the contract shows that the discounting rate measures the aversion to conservation characteristics of the consumer. The analysis of the contract shows that a sufficiently time-patient consumer is not penalized for the deviation, as the consumer himself is aware of conservation requirements. This result is similar to the results for the present-biased preferences in behavioral economics. Lastly, the infinite horizon contract is extended to two consumers case which internalizes the externality a consumer causes to another. In the two consumer case, consumers are strategically noninteracting; and it is shown that the producer acts as a budget balancer. These contracts are also shown to be economically efficient. The demand management contracts achieve both the procedural and end-state fairness. Also, the infinite horizon contracts are homeomorphic to infinite horizon alternating bargaining model. The efficiency of infinite horizon contracts is due to their homeomorphism with the alternating bargaining process as they exhaust all possible mutual gains from exchange. In the two-period model, the bargaining process is constrained and hence all possible mutual gains are not eliminated, leading to the inefficiency. In part II of the thesis, we discuss the notions of exchangeability in the Shapley value. The Shapley value is a probabilistic value for the transferable utility (TU) cooperative games, in which each player subjectively assigns probabilities to the events which define their positions in the game. In this part, the objective have been to explore the aspect of subjective probability which leads to the uniqueness of the Shapley value. This aspect of subjective probability is known as exchangeability. We derive the Shapley value using de Finetti’s theorem. We also show that, in the Shapley value, each player’s prospects of joining a t-player game as the last member of the game is a moment sequence of the uniquely determined uniform distribution. We stress on finite exchangeability; and deduce that, with finite exchangeability, the Shapley value is the only value in which the probability assignment is a unique mixture of independent and identical distributions. It is concluded that, in both the finite and infinite exchangeable cases, the uniqueness of probability assignment in the Shapley value is due to exchangeability and the mixing with the uniform distribution.

Na literatura de controle, os controladores preditivos disponíveis apresentam aplicações limitadas em processos com características integradoras. Uma dessas limitações está relacionada às propriedades estabilizantes que surgem da lei de controle quando o sistema é colocado em malha fechada. O principal objetivo dessa dissertação é desenvolver um algoritmo de controle preditivo, nominalmente estável, baseado num horizonte de predição infinito das saídas do sistema. Para essa finalidade, uma representação existente em variáveis de estado é analisada e estendida para ser capaz de ser aplicável para processos estáveis e integradores, com tempo morto. Dois métodos distintos que garantem a estabilidade de sistemas integradores foram desenvolvidos com um MPC de horizonte infinito. Para o caso de realimentação de estado, a estabilidade nominal do MPC proposto foi provada para os dois métodos e a eficiência dos controladores foi verificada através de exemplos da literatura de controle. Um dos controladores foi também aplicado em um processo de destilação de uma refinaria de petróleo. O algoritmo desenvolvido foi estendido para o caso de realimentação das saídas com a inclusão do filtro de Kalman para a estimativa dos estados do modelo. A aplicação do controlador em um sistema com diferentes tempos mortos entre as variáveis integradoras foi também estudada. / In the control literature, the available predictive controllers show limitations to be applied in processes that have integrating behavior. One of these limitations is related to the stabilizing properties of the resulting control law when the system is in closed loop. The main objective of this work is to develop a nominal stable control algorithm based on infinite prediction of system outputs. For this purpose an existing state space representation is explored and extended in order to be able to be applied to stable and integrating process with dead time. Two distinct methods to assure stability of integrating systems were developed for the infinite horizon MPC. For the case of state feedback, nominal stability of the resulting MPC was proved for the two methods and the efficiency of the controllers was verified through examples of the control literature. One of the controllers was also applied to an industrial distillation process of a petroleum refinery. The developed algorithm was extended to the output feedback operation with the inclusion of a Kalman filter for estimation of the model states. The application of the controller to a system with different dead times between integrating variables was also studied.

Cette thèse contient des contributions originales à la théorie du Contrôle Optimal en temps discret et en horizon infini du point de vue de Pontryagin. Il y a 5 chapitres dans cette thèse. Dans le chapitre 1, nous rappelons des résultats préliminaires sur les espaces de suites à valeur dans et des résultats de Calcul Différentiel. Dans le chapitre 2, nous étudions le problème de Contrôle Optimal, en temps discret et en horizon infini avec la contrainte asymptotique et avec le système autonome. En utilisant la structure d'espace affine de Banach de l'ensemble des suites convergentes vers 0, et la structure d'espace vectoriel de Banach de l'ensemble des suites bornées, nous traduisons ce problème en un problème d'optimisation statique dam des espaces de Banach. Après avoir établi des résultats originaux sur les opérateurs de Nemytskii sur les espaces de suites et après avoir adapté à notre problème un théorème d'existence de multiplicateurs, nous établissons un nouveau principe de Pontryagin faible pour notre problème. Dans le chapitre 3, nous établissons un principe de Pontryagin fort pour les problèmes considérés au chapitre 2 en utilisant un résultat de Ioffe-Tihomirov. Le chapitre 4 est consacré aux problèmes de Contrôle Optimal, en temps discret et en horizon infini, généraux avec plusieurs critères différents. La méthode utilisée est celle de la réduction à l'horizon fini, initiée par J. Blot et H. Chebbi en 2000. Les problèmes considérés sont gouvernés par des équations aux différences ou des inéquations aux différences. Un nouveau principe de Pontryagin faible est établi en utilisant un résultat récent de J. Blot sur les multiplicateurs à la Fritz John. Le chapitre 5 est consacré aux problèmes multicritères de Contrôle Optimal en temps discret et en horizon infini. De nouveaux principes de Pontryagin faibles et forts sont établis, là-aussi en utilisant des résultats récents d'optimisation, sous des hypothèses plus faibles que celles des résultats existants. / This thesis contains original contributions to the optimal control theory in the discrete-time framework and in infinite horizon following the viewpoint of Pontryagin. There are 5 chapters in this thesis. In Chapter 1, we recall preliminary results on sequence spaces and on differential calculus in normed linear space. In Chapter 2, we study a single-objective optimal control problem in discrete-time framework and in infinite horizon with an asymptotic constraint and with autonomous system. We use an approach of functional analytic for this problem after translating it into the form of an optimization problem in Banach (sequence) spaces. Then a weak Pontyagin principle is established for this problem by using a classical multiplier rule in Banach spaces. In Chapter 3, we establish a strong Pontryagin principle for the problems considered in Chapter 2 using a result of Ioffe and Tihomirov. Chapter 4 is devoted to the problems of Optimal Control, in discrete time framework and in infinite horizon, which are more general with several different criteria. The used method is the reduction to finite-horizon initiated by J. Blot and H. Chebbi in 2000. The considered problems are governed by difference equations or difference inequations. A new weak Pontryagin principle is established using a recent result of J. Blot on the Fritz John multipliers. Chapter 5 deals with the multicriteria optimal control problems in discrete time framework and infinite horizon. New weak and strong Pontryagin principles are established, again using recent optimization results, under lighter assumptions than existing ones.

最近，在考慮一個小型開放經濟體系下，貿易條件(Terms of Trade)外生變化對其經 常帳(Current Account) 影響的相關文獻上，以兩期模型(Infinite-horizon Modcl) 處理時，若貿易條件恆常地惡化，則一方面由於實質所得減少造成儲蓄降低，因而使 得經常帳惡化 (此即所謂「財富效果」) ；另一方面則經由實質利率改變影響儲蓄與 投資決策，因此經常帳再度隨之調整 (此稱為「實質利率效果」) 。至於貿易條件惡 化對於經常帳究有改善或惡化的影響則端視各模型處理時假設不同有不同的結論。 在Heckscher-Ohlin 生產技術的假設下，兩要素用在兩部門間生產，會使得貿易條件 透過第三管道影響經常帳。由Stolper-Samuelson 定理可知，貿易條件惡化將會降低 出口財較密集使用要素的報酬，提高進口財較密集使用要素的報酬，在所得重新分配 之後儲蓄決策將有所改變，因而經常帳也跟著受影響。此稱之為 Stolper-Samuelson 效果。 本文擬運用Blanchard 式的跨代模型(Overlapping-generations Mode)考慮一個小型 開放經濟體系在面臨外生貿易條件惡化時，其經常帳與資本累積的動態變化。此模型 假設每個經濟個體(agent) 活有限期，因此長期均衡值不必滿足時間偏好率等於利率 的條件，我們在此考慮投資與儲蓄的動態決策行為。又由於假設Heckscher-Ohlin 生 產技術，本文也將著重於討論Stolper-Samuelson 效果在此模型中的影響。

In this thesis, we study the existence of pathwise random periodic solutions to both the semilinear stochastic differential equations with linear multiplicative noise and the semilinear stochastic partial differential equations with linear multiplicative noise in a Hilbert space. We identify them as the solutions of coupled forward-backward infinite horizon stochastic integral equations in general cases, and then perform the argument of the relative compactness of Wiener-Sobolev spaces in C([0, T],L2Ω,Rd)) or C([0, T],L2(Ω x O)) and Schauder's fixed point theorem to show the existence of a solution of the coupled stochastic forward-backward infinite horizon integral equations.

In Chapter 2, we propose several two-timescale simulation-based actor-critic algorithms for solution of infinite horizon Markov Decision Processes (MDPs) with finite state-space under the average cost criterion. On the slower timescale, all the algorithms perform a gradient search over corresponding policy spaces using two different Simultaneous Perturbation Stochastic Approximation (SPSA) gradient estimates. On the faster timescale, the differential cost function corresponding to a given stationary policy is updated and averaged for enhanced performance. A proof of convergence to a locally optimal policy is presented. Next, a memory efficient implementation using a feature-vector representation of the state-space and TD (0) learning along the faster timescale is discussed. A three-timescale simulation based algorithm for solution of infinite horizon discounted-cost MDPs via the Value Iteration approach is also proposed. An approximation of the Dynamic Programming operator T is applied to the value function iterates. A sketch of convergence explaining the dynamics of the algorithm using associated ODEs is presented. Numerical experiments on rate based flow control on a bottleneck node using a continuous-time queueing model are presented using the proposed algorithms. Next, in Chapter 3, we develop three simulation-based algorithms for finite-horizon MDPs (FHMDPs). The first algorithm is developed for finite state and compact action spaces while the other two are for finite state and finite action spaces. Convergence analysis is briefly sketched. We then concentrate on methods to mitigate the curse of dimensionality that affects FH-MDPs severely, as there is one probability transition matrix per stage. Two parametrized actor-critic algorithms for FHMDPs with compact action sets are proposed, the ‘critic’ in both algorithms learning the policy gradient. We show w.p1convergence to a set with the necessary condition for constrained optima. Further, a third algorithm for stochastic control of stopping time processes is presented. Numerical experiments with the proposed finite-horizon algorithms are shown for a problem of flow control in communication networks. Towards stochastic optimization, in Chapter 4, we propose five algorithms which are variants of SPSA. The original one measurement SPSA uses an estimate of the gradient of objective function L containing an additional bias term not seen in two-measurement SPSA. We propose a one-measurement algorithm that eliminates this bias, and has asymptotic convergence properties making for easier comparison with the two-measurement SPSA. The algorithm, under certain conditions, outperforms both forms of SPSA with the only overhead being the storage of a single measurement. We also propose a similar algorithm that uses perturbations obtained from normalized Hadamard matrices. The convergence w.p.1 of both algorithms is established. We extend measurement reuse to design three second-order SPSA algorithms, sketch the convergence analysis and present simulation results on an illustrative minimization problem. We then propose several stochastic approximation implementations for related algorithms in flow-control of communication networks, beginning with a discrete-time implementation of Kelly’s primal flow-control algorithm. Convergence with probability1 is shown, even in the presence of communication delays and stochastic effects seen in link congestion indications. Two relevant enhancements are then pursued :a) an implementation of the primal algorithm using second-order information, and b) an implementation where edge-routers rectify misbehaving flows. Also, discrete-time implementations of Kelly’s dual algorithm and primal-dual algorithm are proposed. Simulation results a) verifying the proposed algorithms and, b) comparing stability properties with an algorithm in the literature are presented.

Cette thèse traite des phénomènes liés aux mathématiques financières et économiques. Elle est composée de deux sujets de recherche indépendants. La première partie est consacrée à deux contributions au problème de Merton. Pour commencer, nous étudions le problème de l’investissement optimal et de la consommation de Merton dans le cas de marchés discrets dans un horizon infini. Nous supposons qu’il y a des frictions sur les marchés en raison de la perte due aux échanges financières. Ces frictions sont modélisées par des fonctions de pénalités non linéaires où les modèles classiques de coût de transactions étudiés par Magill et Constantinides [31] et les marchés illiquides étudiés par Cetin, Jarrow et Protter dans [6] sont inclus dans cette formulation. Dans ce contexte, la région de solvabilité est définie en tenant compte de cette fonction de pénalité et chaque investisseur doit maximiser son utilité, dérivée de la consommation. Nous donnons la programmation dynamique du modèle et nous prouvons l’existence et l’unicité de la fonction valeur. Des stratégies optimales d’investissement et de consommation sont également construites. Ensuite, nous étendons le modèle de Merton à un problème à plusieurs investisseurs. Notre approche consiste à construire un modèle d’équilibre général déterministe dynamique. Nous prouvons ensuite l’existence d’un équilibre du problème qui est un ensemble de contrôles composés de processus de consommation et de portefeuille, ainsi que les processus de prix qui en découlent afin que la politique de consommation de chaque investisseur maximise son profil. Les résultats obtenus dans cette partie étendent principalement les résultats récemment obtenus par Chebbi et Soner [10] ainsi qu’aux d’autres résultats obtenus dans ce cadre dans la littérature. Dans la deuxième partie, nous traitons le problème de l’existence d’un équilibre d’une économie de production avec des ensembles d’allocations réalisables non-bornés où les consommateurs peuvent avoir des préférences non-transitives non-complètes. Nous introduisons une propriété asymptotique sur les préférences pour les consommations réalisables afin de prouver l’existence d’un équilibre. Nous montrons que cette condition est vraie lorsque l’ensemble des allocations réalisables est compact ou aussi lorsque les préférences sont représentées par des fonctions d’utilité dans le cas où l’ensemble des niveaux d’utilité rationnels individuels réalisables est compact. Cette hypothèse généralise la condition de CPP de Allouch [1] et couvre l’exemple de Page et al. [40] lorsque les niveaux d’utilité disponibles définis ne sont pas compacts. Nous étendons donc les résultats existants dans la littérature avec des ensembles réalisables non bornés de deux façons en ajoutant la production et en prenant en compte des préférences générales. / This PhD dissertation studies two independent research topics dealing with phenomena issues from financial and economic mathematics.This thesis is organized in two parts. The first part is devoted to two contributions tothe Merton problem. First, we investigate the problem of optimal investment and consumption of Merton in the case of discrete markets in an infinite horizon. We suppose that there is frictions in the markets due to loss in trading. These frictions are modeled through nonlinear penalty functions and the classical transaction cost studied by Magill and Constantinides in [31] and illiquidity models studied by Cetin, Jarrow and Protter in [6] are included in this formulation. In this context, the solvency region is defined taking into account this penalty function and every investigator have to maximize his utility, that is derived from consumption, in this region. We give the dynamic programming ofthe model and we prove the existence and uniqueness of the value function. Optimalinvestment and consumption strategies are constructed as well. We second extend the Merton model to a multi-investors problem. Our approach is to construct a dynamic deterministic general equilibrium model. We then provide the existence of equilibrium of the problem which is a set of controls that is composed of consumption and portfolio processes, as well as the resulting price processes so that each investor’s consumption policy maximizes his lifetime expected. The results obtained in this part extends mainly the results recently obtained by Chebbi and Soner [10] and other corresponding results in the litterature.The second part of this thesis deals with the problem of the existence of an equilibrium of a production economy with unbounded attainable allocations sets where the consumers may have non-complete non-transitive preferences. We introduce an asymptotic property on preferences for the attainable consumptions in order to prove the existence of an equilibrium. We show that this condition holds true if the set of attainable allocations is compact or, when preferences are representable by utility functions, if the set of attainable individually rational utility levels is compact. This assumption generalizes the CPP condition of Allouch [1] and covers the example of Page et al. [40] when the attainable utility levels set is not compact. So we extend the previous existence results with unbounded attainable sets in two ways by adding a production sector and considering general preferences.