This dissertation consists of three essays in Economic Theory. The rst essay proposes and studies a new solution concept for games with incomplete information. In game theory, there is a basic methodological dichotomy between Harsanyi's \game-theoretic" view and Aumann's \Bayesian decision-theoretic" view of the world. We follow the game theoretic view, propose and study interim partially correlated rationalizability for games with incomplete information. We argue that the distinction between this solution concept and the interim correlated rationalizability studied by Dekel, Fudenberg and Morris (2007) is fundamental, in that the latter implicitly follows Aumann's Bayesian view. Our main result shows that two types provide the same prediction in interim partially correlated rationalizability if and only if they have the same in nite hierarchy of beliefs over conditional beliefs. We also establish an equivalence result between this solution concept and the Bayesian solution{a notion of correlated equilibrium proposed by Forges (1993). The second essay studies the relationship between correlated equilibrium the redundancy embedded in type spaces. The Bayesian solution is a notion of correlated equilibrium proposed by Forges (1993), and hierarchies of beliefs over conditional beliefs are introduced by Ely and Peski (2006) in their study of interim rationalizability. We study the connection between the two concepts. We say that two type spaces are equivalent if they represent the same set of hierarchies of beliefs over conditional beliefs. We show that the correlation embedded in equivalent type spaces can be characterized by partially correlating devices, which send correlated signals to players in a belief invariant way. Since such correlating devices also implement the Bayesian solution, we establish that the Bayesian solution is invariant across equivalent type spaces. The third essay studies the existence of equilibria for rst-price sealed bid auctions when bidders form a network and each bidder observes perfectly their neighbors' private valuations. Asymmetry in bidders' positions in the network creates asymmetry in bidders' knowledge. We show the existence of pure-strategy equilibrium. / text

L'objectif de cette thèse est l'étude des jeux différentiels stochastiques à information incomplète. Nous considérons un jeu à deux joueurs adverses qui contrôlent une diffusion afin de minimiser, respectivement de maximiser un paiement spécifique. Pour modéliser l'incomplétude des informations, nous suivrons la célèbre approche d'Aumann et Maschler. Nous supposons qu'il existe des états de la nature différents dans laquelle le jeu peut avoir lieu. Avant que le jeu commence, l'état est choisi au hasard. L'information est ensuite transmise à un joueur alors que le second ne connaît que les probabilités respectives pour chaque état.Dans cette thèse nous établissons une représentationduale pour les jeux différentiels stochastiques à information incomplète. Ici, nous utilisons largement la théorie des équations différentielles stochastiques rétrogrades (EDSRs), qui se révèle être un outilindispensable dans cette étude. En outre, nous montrons comment, sous certaines restrictions, cette représentation permetde construire des stratégies optimales pour le joueur informé. Ensuite, nous donnons, en utilisant la représentation duale, une preuve particulièrement simple de la semiconvexité de la fonction valeur des jeux différentiels à information incomplète.Un autre partie de la thèse est consacré à des schémas numériques pour les jeux différentiels stochastiques à informationincomplète. Dans la dernière partie nous étudions des jeux d'arrêt optimal en temps continue, appelés jeux de Dynkin, à information incomplète. Nous établissons également une représentation duale, qui est utilisé pour déterminer des stratégies optimales pour le joueur informé dans ce cas. / The objective of this thesis is the study of stochastic differential games with incomplete information. We consider a game with two opponent players who control a diffusion in order to minimize, respectively maximize a certain payoff. To model the information incompleteness we will follow the famous ansatz of Aumann and Maschler. We assume that there are different states of nature in which the game can take place. Before the game starts the state is chosen randomly. The information is then transmitted to one player while the second one only knows the respective probabilities for each state. In this thesis we establish a dual representation for stochastic differential games with incomplete information. Therein we make a vast use of the theory of backward stochastic differential equations (BSDEs), which turns out to be an indispensable tool in this study. Moreover we show how under some restrictions that this representation allows to construct optimal strategies for the informed player.Morover we give - using the dual representation - a strikingly simple proof for semiconvexity of the value function of differential games with incomplete information. Another part of this thesis is devoted to numerical schemes for stochastic differential games with incomplete information. In the last part we investigate continuous time optimal stopping games, so called Dynkin games, with information incompleteness. We show that these games have a value and a unique characterization by a fully non-linear variational PDE for which we provide a comparison principle. Also we establish a dual representation for Dynkin games with incomplete information.