Search

International audience; In this paper we obtain a Wong-Zakai approximation to solutions of backward doubly stochastic differential equations.

We prove that there exists a functorial correspondence between MV-algebras and partially cyclically ordered groups which are wound round of lattice-ordered groups. It follows that some results about cyclically ordered groups can be stated in terms of MV-algebras. For example, the study of groups together with a cyclic order allows to get a first-order characterization of groups of unimodular complex numbers and of finite cyclic groups. We deduce a characterization of pseudofinite MV-chains and of pseudo-simple MV-chains (i.e. which share the same first-order properties as some simple ones). We can generalize these results to some non-lineraly ordered MV-algebras, for example hyper-archimedean MV-algebras.

We address a general optimal switching problem over finite horizon for a stochastic system described by a differential equation driven by Brownian motion. The main novelty is the fact that we allow for infinitely many modes (or regimes, i.e. the possible values of the piecewise-constant control process). We allow all the given coefficients in the model to be path-dependent, that is, their value at any time depends on the past trajectory of the controlled system. The main aim is to introduce a suitable (scalar) backward stochastic differential equation (BSDE), with a constraint on the martingale part, that allows to give a probabilistic representation of the value function of the given problem. This is achieved by randomization of control, i.e. by introducing an auxiliary optimization problem which has the same value as the starting optimal switching problem and for which the desired BSDE representation is obtained. In comparison with the existing literature we do not rely on a system of reflected BSDE nor can we use the associated Hamilton-Jacobi-Bellman equation in our non-Markovian framework., Comment: 36 pages

It is considered the problem of localization on the plane of two radioactive sources by K detectors. Each detector records a realization of inhomogeneous Poisson process and the intensity function of this process is a sum of a signal arriving from the sources and the constant Poisson noise of known intensity. The time of the beginning of emissions of two sources is known and the main problem is the estimation of the position of the sources. The properties of the MLE and Bayessian estimators are described in the asymptotics of large signals in three situations of different regularities of the fronts of the signals: smooth, cusp-type and change-point type.

The ghost note is a natural note which can be played exclusively on bass brass instruments with a predominantly-expanding bore profile such as tubas, euphoniums or saxhorns. It stands between the pedal note-the lowest natural note playable, or first regime-and the instrument's second regime. However, if the interval between the pedal note and the second regime remains close to an octave regardless of the instrument, the interval between the pedal note and the ghost note vary from a minor third to a perfect fourth. References about this note are very scarce, and it is not commonly known among tuba players.This study shows that an elementary brass model describing the player coupled to the instrument is capable of bringing both the ghost and the pedal note to light. Here, we adopt a dynamical systems point of view and perform a bifurcation analysis using a software of numerical continuation. The numerical results provided in terms of frequency intervals between pedal note and ghost note are compared with frequency intervals experimentally inferred from recordings of seven different types of tuba, each of them being played by two professional tuba players., Comment: arXiv admin note: text overlap with arXiv:2112.08751

In this study, an acoustic resonator – a bass brass instrument – with multiple resonances coupled to an exciter – the player’s lips – with one resonance is modelled by a multidimensional dynamical system, and studied using a continuation and bifurcation software. Bifurcation diagrams are explored with respect to the blowing pressure, in particular with focus on the minimal blowing pressure allowing stable periodic oscillations and the associated frequency. The behaviour of the instrument is first studied close to a (non oscillating) equilibrium using linear stability analysis. This allows to determine the conditions at which an equilibrium destabilises and as such where oscillating regimes can emerge (corresponding to a sound production). This approach is useful to characterise the ease of playing of a brass instrument, which is assumed here to be related – as a first approximation – to the linear threshold pressure. In particular, the lower the threshold pressure, the lower the physical effort the player has to make to play a note [The Science of Brass Instruments. Springer-Verlag, 2021]. Cases are highlighted where periodic solutions in the bifurcation diagrams are reached for blowing pressures below the value given by the linear stability analysis. Thus, bifurcation diagrams allow a more in-depth analysis. Particular attention is devoted to the first playing regime of bass brass instruments (the pedal note and the ghost note of a tuba in particular), whose behaviour qualitatively differs from a trombone to a euphonium for instance.

We make available some results about model theory cyclically ordered groups. We start with a classification of complete theories of divisible abelian cyclically ordered groups. Most of this paper is dedicated to the study of the model theory of torsion-free divisible abelian cyclically ordered groups, since we first show that the theory of divisible abelian cyclically ordered groups with non-trivial torsion elements is complete. Then we look at the cyclically ordered groups where the only parametrically definable subsets are finite unions of singletons and open intervals, and those where the definable subsets are finite union of singletons and c-convex subsets, where being c-convex is the analogue of being convex in the linearly ordered case.

The aim of this short note is to fill in a gap in our earlier paper [16] on 2BSDEs with reflections, and to explain how to correct the subsequent results in the second paper [15]. We also provide more insight on the properties of 2RBSDEs, in the light of the recent contributions [13, 23] in the so--called $G-$framework., 16 pages

This paper investigates the Hamilton-Jacobi-Bellman system of equations associated with the m-states optimal switching problem in finite horizon when the state process lives in a connected bounded closed domain. We show existence and uniqueness of the solution in viscosity sense of the system. We use systems of generalized reflected backward stochastic differential equations with oblique reflection and the Feynman-Kac representation of their solutions in the Markovian framework.

We consider a mean field game (MFG) of optimal portfolio liquidation under asymmetric information. We prove that the solution to the MFG can be characterized in terms of a forward-backward stochastic differential equation (FBSDE) with a possibly singular terminal condition on the backward component or, equivalently, in terms of an FBSDE with a finite terminal value yet a singular driver. Extending the method of continuation to linear-quadratic FBSDEs with a singular driver, we prove that the MFG has a unique solution. Our existence and uniqueness result allows proving that the MFG with a possibly singular terminal condition can be approximated by a sequence of MFGs with finite terminal values.

This paper considers the joint estimation of the parameters of a first-order fractional autoregressive model by constructing an initial estimator with convergence speed lower than √ n and singular asymptotic joint distribution. The one-step procedure is then used in order to obtain an asymptoticallyefficient estimator. This estimator is computed faster than the maximum likelihood or Whittle estimator and therefore allows for faster inference on large samples. The paper illustrates the performance of this method on finite-size samples via Monte Carlo simulations.

International audience

We prove an existence and uniqueness result for two-obstacle problem for quasilinear Stochastic PDEs (DOSPDEs for short). The method is based on the probabilistic interpretation of the solution by using the backward doubly stochastic differential equations (BDSDEs for short).

International audience; Fundamental research on ageing has taken an interesting turn in recent years with the rapid development of biomarkers predicting mortality in model organisms, particularly Drosophila, as well as in humans through improvements in approaches to the identification of circulating molecules in mass. These developments lead to a shift in our ability to predict the occurrence of death from the historically population level to the individual level. We question here the ethical, medical and social implications of this change of scale.; La recherche fondamentale sur le vieillissement a pris un tour intéressant ces dernières années avec un développement rapide des biomarqueurs prédictifs de mortalité chez les organismes modèles, notamment la drosophile, ainsi que chez l’être humain à travers les améliorations des approches d’identification en masse de molécules circulantes. Ces développements conduisent à un déplacement de notre capacité de prédiction de survenue de la mort, du niveau historiquement populationnel au niveau individuel. Nous interrogeons ici les implications éthiques, médicales et sociales de ce changement d’échelle.

Fundamental research on ageing has taken an interesting turn in recent years with the rapid development of biomarkers predicting mortality in model organisms, particularly Drosophila, as well as in humans through improvements in approaches to the identification of circulating molecules in mass. These developments lead to a shift in our ability to predict the occurrence of death from the historically population level to the individual level. We question here the ethical, medical and social implications of this change of scale., La recherche fondamentale sur le vieillissement a pris un tour intéressant ces dernières années avec un développement rapide des biomarqueurs prédictifs de mortalité chez les organismes modèles, notamment la drosophile, ainsi que chez l’être humain à travers les améliorations des approches d’identification en masse de molécules circulantes. Ces développements conduisent à un déplacement de notre capacité de prédiction de survenue de la mort, du niveau historiquement populationnel au niveau individuel. Nous interrogeons ici les implications éthiques, médicales et sociales de ce changement d’échelle.

We use the functional Itô calculus to prove that the solution of a BSDE with singular terminal condition verifies at the terminal time: $\liminf _{t\to T} Y(t) = \xi = Y(T)$liminft→TY(t)=ξ=Y(T). Hence, we extend known results for a non-Markovian terminal condition.

This thesis is devoted to asymptotic inferenre of differents chronological models driven by a noise with memory. In these models, the least squares estimator is not consistent and we consider other estimators. We begin by studying the almost-sureasymptotic properties of the maximum likelihood estimator of the autoregressive coefficient in an autoregressive process drivenby a stationary Gaussian noise. We then present a statistical procedure in order to detect a change of regime within this model,taking inspiration from the classic case driven by a strong white noise. Then we consider an autoregressive model where the coefficients are random and have a short memory. Here again, the least squares estimator is not consistent and we correct the previous statistic in order to correctly estimate the parameters of the model. Finally we study a new joint estimator of the Hurst exponent and the variance in a fractional Gaussian noise observed at high frequency whose qualities are comparable to the maximum likelihood estimator.; Cette thèse est dévolue à la statistique inférentielle asymptotique de différents modèles chronologiques dirigés par un bruit comportant de la mémoire. Dans ces modèles, l'estimateur des moindres carrés n'est pas consistant et nous considérons d'autres estimateurs. Nous commençons par étudier les propriétés asymptotiques presquesûres de l'estimateur du maximum de vraisemblance du coefficient d'autorégression dans un processus autorégressif dirigé par un bruit gaussien stationnaire. Nous présentons ensuite une procédure statistique afin de détecter un changement de régime au sein de ce modèle en s'inspirant du cas classique dirigé par un bruit blanc fort. Nous abordons ensuite un modèle autorégressif où les coefficients sont aléatoires et possèdent une courte mémoire. Là encore l'estimateur des moindres carrés n'est pas consistant et nouscorrigeons l'estimation afin d'estimer correctement les paramètres du modèle. Pour finir nous étudions un nouvel estimateur joint de l'exposant de Hurst et de la variance dans un bruit gaussien fractionnaire observé à haute fréquence dont les qualités sont comparables au maximum de vraisemblance.

This thesis is devoted to asymptotic inferenre of differents chronological models driven by a noise with memory. In these models, the least squares estimator is not consistent and we consider other estimators. We begin by studying the almost-sureasymptotic properties of the maximum likelihood estimator of the autoregressive coefficient in an autoregressive process drivenby a stationary Gaussian noise. We then present a statistical procedure in order to detect a change of regime within this model,taking inspiration from the classic case driven by a strong white noise. Then we consider an autoregressive model where the coefficients are random and have a short memory. Here again, the least squares estimator is not consistent and we correct the previous statistic in order to correctly estimate the parameters of the model. Finally we study a new joint estimator of the Hurst exponent and the variance in a fractional Gaussian noise observed at high frequency whose qualities are comparable to the maximum likelihood estimator.; Cette thèse est dévolue à la statistique inférentielle asymptotique de différents modèles chronologiques dirigés par un bruit comportant de la mémoire. Dans ces modèles, l'estimateur des moindres carrés n'est pas consistant et nous considérons d'autres estimateurs. Nous commençons par étudier les propriétés asymptotiques presquesûres de l'estimateur du maximum de vraisemblance du coefficient d'autorégression dans un processus autorégressif dirigé par un bruit gaussien stationnaire. Nous présentons ensuite une procédure statistique afin de détecter un changement de régime au sein de ce modèle en s'inspirant du cas classique dirigé par un bruit blanc fort. Nous abordons ensuite un modèle autorégressif où les coefficients sont aléatoires et possèdent une courte mémoire. Là encore l'estimateur des moindres carrés n'est pas consistant et nouscorrigeons l'estimation afin d'estimer correctement les paramètres du modèle. Pour finir nous étudions un nouvel estimateur joint de l'exposant de Hurst et de la variance dans un bruit gaussien fractionnaire observé à haute fréquence dont les qualités sont comparables au maximum de vraisemblance.

We consider Cauchy problem for a divergence form second order parabolic operator with rapidly oscillating coefficients that are periodic in spatial variable and random stationary ergodic in time. As was proved in [25] and [13] in this case the homogenized operator is deterministic. We obtain the leading terms of the asymptotic expansion of the solution , these terms being deterministic functions, and show that a properly renormalized difference between the solution and the said leading terms converges to a solution of some SPDE.

This paper is related to the study of systems of reflected backward stochastic differential equations with interconnected bilateral obstacles. These systems are connected with zero-sum stochastic switching games. Under appropriate assumptions, we provide either existence or existence and uniqueness of the solution of those systems when the switching costs are Ito processes. The link with systems of PDEs with bilateral interconnected obstacles is also stated via the Feynman-Kac representation when randomness comes from a Markov diffusion process.

In this paper we study a zero-sum switching game and its verification theorems expressed in terms of either a system of Reflected Backward Stochastic Differential Equations (RBSDEs in short) with bilateral interconnected obstacles or a system of parabolic partial differential equations (PDEs in short) with bilateral interconnected obstacles as well. We show that each one of the systems has a unique solution. Then we show that the game has a value.

International audience; We consider an RCAR(p) process and we establish that the standard estimation lacks consistency as soon as there exists a nonzero serial correlation in the coefficients. We give the correct asymptotic behavior and some simulations come to illustrate the results

This paper addresses the problem of estimating drift parameter of the Ornstein - Uhlenbeck type process, driven by the sum of independent standard and fractional Brownian motions. The maximum likelihood estimator is shown to be consistent and asymptotically normal in the large-sample limit, using some recent results on the canonical representation and spectral structure of mixed processes., Comment: to appear in Theory of Probability and its Applications

We consider a variant of the basic problem of the calculus of variations, where the Lagrangian is convex and subject to randomness adapted to a Brownian filtration. We solve the problem by reducing it, via a limiting argument, to an unconstrained control problem that consists in finding an absolutely continuous process minimizing the expected sum of the Lagrangian and the deviation of the terminal state from a given target position. Using the Pontryagin maximum principle, we characterize a solution of the unconstrained control problem in terms of a fully coupled forward–backward stochastic differential equation (FBSDE). We use the method of decoupling fields for proving that the FBSDE has a unique solution. We exploit a monotonicity property of the decoupling field for solving the original constrained problem and characterize its solution in terms of an FBSDE with a free backward part.

In this paper, we provide a one-to-one correspondence between the solution Y of a BSDE with singular terminal condition and the solution H of a BSDE with singular generator. This result provides the precise asymptotic behaviour of Y close to the final time and enlarges the uniqueness result to a wider class of generators.

In this paper, we study backward stochastic Volterra integral equations introduced in Lin [Stochastic Anal. Appl.20(2002) 165–183] and Yong [Stochastic Process. Appl.116(2006) 779–795] and extend the existence, uniqueness or comparison results for general filtration as in Papapantoleonet al.[Electron. J. Probab.23(2018) EJP240] (not only Brownian-Poisson setting). We also considerLp-data and explore the time regularity of the solution in the Itô setting, which is also new in this jump setting.

Even on modern straight pianos, the inharmonicity of the lower strings is rather large especially for the first octave. Consequently, the timber of these strings can sometimes sound awful and chords on the first octave be highly dissonant. The idea of the present study is to show how this defect can be rectified using an in-homogeneous winding on the whole string in order to minimize inharmonicity. The problem is solved using an optimisation procedure considering a non uniform linear density. Results show that the inharmonicity of the first partials could be highly reduced by a non uniform winding limited to a quarter of the string., Sur les pianos droits l'inharmonicité des cordes graves est assez grande, surtout pour la première octave. Par conséquent, le son de ces cordes peut parfois sembler hideux et les accords sur la première octave sont très dissonants. L’idée de la présente étude est de montrer comment remédier à ce défaut en utilisant un filage inhomogène sur l’ensemble de la corde afin d'en minimiser l’inharmonicité. Le problème est résolu en utilisant une procédure d'optimisation prenant en compte une densité linéaire non uniforme. Les résultats montrent que l’inharmonicité des premiers partiels pourrait être fortement réduite par un filage non uniforme limité au quart de la corde.

Even on modern straight pianos, the inharmonicity of the lower strings is rather large especially for the first octave. Consequently, the timber of these strings can sometimes sound awful and chords on the first octave be highly dissonant. The idea of the present study is to show how this defect can be rectified using an in-homogeneous winding on the whole string in order to minimize inharmonicity. The problem is solved using an optimisation procedure considering a non uniform linear density. Results show that the inharmonicity of the first partials could be highly reduced by a non uniform winding limited to a quarter of the string.; Sur les pianos droits l'inharmonicité des cordes graves est assez grande, surtout pour la première octave. Par conséquent, le son de ces cordes peut parfois sembler hideux et les accords sur la première octave sont très dissonants. L’idée de la présente étude est de montrer comment remédier à ce défaut en utilisant un filage inhomogène sur l’ensemble de la corde afin d'en minimiser l’inharmonicité. Le problème est résolu en utilisant une procédure d'optimisation prenant en compte une densité linéaire non uniforme. Les résultats montrent que l’inharmonicité des premiers partiels pourrait être fortement réduite par un filage non uniforme limité au quart de la corde.

This paper investigates two existence theorems for the path-dependent heat equation, which is the Kolmogorov equation related to the window Brownian motion, considered as a C ⁢ ( [ - T , 0 ] ) {C([-T,0])} -valued process. We concentrate on two general existence results of its classical solutions related to different classes of terminal conditions: the first one is given by a cylindrical not necessarily smooth random variable, the second one is a smooth generic functional.

A random coefficient autoregressive process is deeply investigated in which the coefficients are correlated. First we look at the existence of a strictly stationary causal solution, we give the second-order stationarity conditions and the autocorrelation function of the process. Then we study some asymptotic properties of the empirical mean and the usual estimators of the process, such as convergence, asymptotic normality and rates of convergence, supplied with the appropriate assumptions on the driving perturbations. Our objective is to get an overview of the influence of correlated coefficients in the estimation step, through a simple model. In particular, the lack of consistency is shown for the estimation of the autoregressive parameter when the independence hypothesis is violated in the random coefficients. Finally, a consistent estimation is given together with a testing procedure for the existence of correlation in the coefficients. While convergence properties rely on the ergodicity, we use a martingale approach to reach most of the results., Comment: 29 pages

We study a class of nonlinear BSDEs with a superlinear driver process f adapted to a filtration F and over a random time interval [[0, S]] where S is a stopping time of F. The terminal condition $\xi$ is allowed to take the value +$\infty$, i.e., singular. Our goal is to show existence of solutions to the BSDE in this setting. We will do so by proving that the minimal supersolution to the BSDE is a solution, i.e., attains the terminal values with probability 1. We consider three types of terminal values: 1) Markovian: i.e., $\xi$ is of the form $\xi$ = g($\Xi$ S) where $\Xi$ is a continuous Markovian diffusion process and S is a hitting time of $\Xi$ and g is a deterministic function 2) terminal conditions of the form $\xi$ = $\infty$ $\times$ 1 {$\tau$ $\le$S} and 3) $\xi$ 2 = $\infty$ $\times$ 1 {$\tau$ >S} where $\tau$ is another stopping time. For general $\xi$ we prove the minimal supersolution is continuous at time S provided that F is left continuous at time S. We call a stopping time S solvable with respect to a given BSDE and filtration if the BSDE has a minimal supersolution with terminal value $\infty$ at terminal time S. The concept of solvability plays a key role in many of the arguments. Finally, we discuss implications of our results on the Markovian terminal conditions to solution of nonlinear elliptic PDE with singular boundary conditions.

Any dynamic or stochastic notion of a general equilibrium relies on the underlying commodity space. Under sole risk and without multiple–prior uncertainty the usual choice is a Lebesgue space from standard measure theory. In the case of volatility uncertainty it turns out that such a type of function space is no longer appropriate. For this reason we introduce and discuss a new natural commodity space, which can be constructed in three independent and equivalent ways. Each approach departs from one possible way to construct Lebesgue spaces. Moreover, we give a complete representation of the resulting topological dual space. This extends the classic Riesz representation in a natural way. Elements therein are the candidates for a linear equilibrium price system. This representation result has direct implications for the microeconomic foundation of finance under Knightian uncertainty.

International audience; The present paper concerns the parametric estimation for the fractional Gaussian noise in a high-frequency observation scheme. The sequence of Le Cam’s one-step maximum likelihood estimators (OSMLE) is studied. This sequence is defined by an initial sequence of quadratic generalized variations-based estimators (QGV) and a single Fisher scoring step. The sequence of OSMLE is proved to be asymptotically efficient as the sequence of maximum likelihood estimators but is much less computationally demanding. It is also advantageous with respect to the QGV which is not variance efficient. Performances of the estimators on finite size observation samples are illustrated by means of Monte-Carlo simulations.

International audience; A reed instrument model with N acoustical modes can be described as a 2N dimensional autonomous nonlinear dynamical system. Here, a simplied model of a reed-like instrument having two quasi-harmonic resonances, represented by a 4 dimensional dynamical system, is studied using the continuation and bifurcation software AUTO. Bifurcation diagrams of equilibria and periodic solutions are explored with respect to the blowing mouth pressure, with focus on amplitude and frequency evolutions along the dierent solution branches. Equilibria and periodic regimes are connected through Hopf bifurcations, which are found to be direct or inverse depending on the physical parameters values. Emerging periodic regimes mainly supported by either the rst acoustic resonance (rst register) or the second acoustic resonance (second register) are successfully identied by the model. An additional periodic branch is also found to emerge from the branch of the second register through a period-doubling bifurcation. The evolution of the oscillation frequency along each branch of the periodic regimes is also predicted by the continuation method. Stability along each branch is computed as well. Some of the results are interpreted in terms of the ease of playing of the reed instrument. The eect of the inharmonicity between the rst two impedance peaks is observed both when the amplitude of the rst is greater than the second, as well as the inverse case. In both cases, the blowing pressure that results in periodic oscillations has a lowest value when the two resonances are harmonic, a theoretical illustration of the Bouasse-Benade prescription.

In this paper, we study a doubly Reflected Backward Stochastic Differential Equation with Jumps (DRBSDEs in short) when the driver have general quadratic growth. We extend the result of Essaky and Hassani [14] to the jump setting and a generator with general exponential quadratic growth.

We consider a class of Backward Stochastic Differential Equations with superlinear driver process $f$ adapted to a filtration supporting at least a $d$ dimensional Brownian motion and a Poisson random measure on ${\mathbb R}^m- \{0\}.$ We consider the following class of terminal conditions $\xi_1 = \infty \cdot 1_{\{\tau_1 \le T\}}$ where $\tau_1$ is any stopping time with a bounded density in a neighborhood of $T$ and $\xi_2 = \infty \cdot 1_{A_T}$ where $A_t$, $t \in [0,T]$ is a decreasing sequence of events adapted to the filtration ${\mathcal F}_t$ that is continuous in probability at $T$. A special case for $\xi_2$ is $A_T = \{\tau_2 > T\}$ where $\tau_2$ is any stopping time such that $P(\tau_2 =T) =0.$ In this setting we prove that the minimal supersolutions of the BSDE are in fact solutions, i.e., they attain almost surely their terminal values. We further show that the first exit time from a time varying domain of a $d$-dimensional diffusion process driven by the Brownian motion with strongly elliptic covariance matrix does have a continuous density; therefore such exit times can be used as $\tau_1$ and $\tau_2$ to define the terminal conditions $\xi_1$ and $\xi_2.$ The proof of existence of the density is based on the classical Green's functions for the associated PDE., Comment: 26 pages, supported by TUBITAK Grant no 118F163

This thesis focuses on backward stochastic differential equation with jumps and their applications. In the first chapter, we study a backward stochastic differential equation (BSDE for short) driven jointly by a Brownian motion and an integer valued random measure that may have infinite activity with compensator being possibly time inhomogeneous. In particular, we are concerned with the case where the driver has quadratic growth and unbounded terminal condition. The existence and uniqueness of the solution are proven by combining a monotone approximation technics and a forward approach. Chapter 2 is devoted to the well-posedness of generalized doubly reflected BSDEs (GDRBSDE for short) with jumps under weaker assumptions on the data. In particular, we study the existence of a solution for a one-dimensional GDRBSDE with jumps when the terminal condition is only measurable with respect to the related filtration and when the coefficient has general stochastic quadratic growth. We also show, in a suitable framework, the connection between our class of backward stochastic differential equations and risk sensitive zero-sum game. In chapter 3, we investigate a general class of fully coupled mean field forward-backward under weak monotonicity conditions without assuming any non-degeneracy assumption on the forward equation. We derive existence and uniqueness results under two different sets of conditions based on proximation schema weither on the forward or the backward equation. Later, we give an application for storage in smart grids.; Cette thèse porte sur l'étude des équations différentielles stochastiques rétrogrades (EDSR) avec sauts et leurs applications.Dans le chapitre 1, nous étudions une classe d'EDSR lorsque le bruit provient d'un mouvement Brownien et d'une mesure aléatoire de saut indépendante à activité infinie. Plus précisément, nous traitons le cas où le générateur est à croissance quadratique et la condition terminale est non bornée. L'existence et l'unicité de la solution sont prouvées en combinant à la fois la procédure d'approximation monotone et une approche progressive. Cette méthode permet de résoudre le cas où la condition terminale est non bornée.Le chapitre 2 est consacré aux EDSR avec sauts généralisées doublement réfléchies sous des hypothèses d’intégrabilités faibles. Plus précisément, on montre l'existence d'une solution pour un générateur à croissance quadratique stochastique et une condition terminale non bornée. Nous montrons également, dans un cadre approprié, la connexion entre notre classe d’équations différentielles stochastiques rétrogrades et les jeu à somme nuls.Dans le chapitre 3, nous considérons une classe générale d'EDSR progressive-rétrograde couplée avec sauts de type Mackean Vlasov sous une condition faible de monotonicité. Les résultats d'existence et d'unicité sont établis sous deux classes d'hypothèses en se basant sur des schémas de perturbations soit de l’équation différentielle stochastique progressive, soit de l’équation différentielle stochastique rétrograde. On conclut le chapitre par un problème de stockage optimal d’énergie dans un parc électrique de type champs moyen.

This thesis is dedicated to the study of backward stochastic differential equations (BSDEs) and their applications. In chapter 1, we study the problem of maximizing the utility from terminal wealth where the stock price may jump and there are investment constraints on the agent 's strategies. We focus on the BSDE whose solution represents the maximal utility, which allows transferring results on quadratic BSDEs, in particular the stability results, to the problem of utility maximisation. In chapter 2, we consider the problem of pricing American options from theoretical and numerical sides based upon an alternative representation of the value of the option in the form of a viscosity solution of a parabolic equation with a nonlinear reaction term. We extend the viscosity solution characterization proved in [Benth, Karlsen and Reikvam 2003] for call/put American option prices to the case of a general payoff function in a multi-dimensional setting. We address two new numerical schemes inspired by the branching processes. Our numerical experiments show that approximating the discontinuous driver of the associated reaction/diffusion PDE by local polynomials is not efficient, while a simple randomization procedure provides very good results. In chapter 3, we prove existence and uniqueness results for a general class of coupled mean-field forward-backward SDEs with jumps under weak monotonicity conditions and without the non-degeneracy assumption on the forward equation and we give an application in the field of storage in smart grids in the case where the production of electricity is unpredictable.; Cette thèse est consacrée à l'étude des équations différentielles stochastiques rétrogrades (EDSR) et leurs applications. Dans le chapitre 1, on étudie le problème de maximisation de l'utilité de la richesse terminale où le prix de l'actif peut être discontinue sous des contraintes sur les stratégies de l'agent. Nous nous concentrons sur l'EDSR dont la solution représente l'utilité maximale, ce qui permet de transférer des résultats sur les EDSR quadratiques, en particulier les résultats de stabilité, au problème de maximisation d'utilité. Dans le chapitre 2, nous considèrons le problème de valorisation d'options Américaines des points de vue théorique et numérique en s'appuyant sur la représentation du prix de l'option comme solution de viscosité d'une équation parabolique non linéaire. Nous étendons le résultat prouvé dans [Benth, Karlsen and Reikvam 2003] pour un put ou call Américain à un cas plus général dans un cadre multidimensionnel. Nous proposons deux schémas numériques inspirés par les processus de branchement. Nos expériences numériques montrent que l'approximation du générateur discontinu, associé à l'EDP, par des polynômes locaux n'est pas efficace tandis qu'une simple procédure de randomisation donne de très bon résultats. Dans le chapitre 3, nous prouvons des résultats d'existence et d'unicité pour une classe générale d'équations progressives-rétrogrades à champs moyen sous une condition de monotonicité faible et une hypothèse non-dégénérescence sur l'équation progressive et nous donnons une application dans le domaine de stockage d'énergie dans le cas où la production d'électricité est imprévisible.

This thesis is dedicated to the study of backward stochastic differential equations (BSDEs) and their applications. In chapter 1, we study the problem of maximizing the utility from terminal wealth where the stock price may jump and there are investment constraints on the agent 's strategies. We focus on the BSDE whose solution represents the maximal utility , which allows transferring results on quadratic BSDEs, in particular the stability results, to the problem of utility maximisation. In chapter 2, we consider the problem of pricing American options from theoretical and numerical sides based upon an alternative representation of the value of the option in the form of a viscosity solution of a parabolic equation with a nonlinear reaction term. We extend the viscosity solution characterization proved in [Benth, Karlsen and Reikvam 2003] for call/put American option prices to the case of a general payoff function in a multi-dimensional setting . We address two new numerical schemes inspired by the branching processes. Our numerical experiments show that approximating the discontinuous driver of the associated reaction/diffusion PDE by local polynomials is not efficient, while a simple randomization procedure provides very good results. In chapter 3, we prove existence and uniqueness results for a general class of coupled mean-field forward-backward SDEs with jumps under weak monotonicity conditions and without the non-degeneracy assumption on the forward equation and we give an application in the field of storage in smart grids in the case where the production of electricity is unpredictable.; Cette thèse est consacrée à l'étude des équations différentielles stochastiques rétrogrades (EDSR) et leurs applications. Dans le chapitre 1, on étudie le problème de maximisation de l'utilité de la richesse terminale où le prix de l'actif peut être discontinue sous des contraintes sur les stratégies de l'agent. Nous nous concentrons sur l'EDSR dont la solution représente l'utilité maximale, ce qui permet de transférer des résultat sur les EDSR quadratiques, en particulier les résultats de stabilité, au problème de maximisation d'utilité. Dans le chapitre 2, nous considèrons le problème de valorisation d'options Américaines des points de vue théorique et numérique en s'appuyant sur la représentation du prix de l'option comme solution de viscosité d'une équation parabolique non linéaire. Nous étendons le résultat prouvé dans [Benth, Karlsen and Reikvam 2003] pour un put ou call Américain à un cas plus général dans un cadre multidimensionnel. Nous proposons deux schémas numériques inspirés par les processus de branchement. Nos expériences numériques montrent que l'approximation du générateur discontinu, associé à l'EDP, par des polynômes locaux n'est pas efficace tandis qu'une simple procédure de randomisation donne de très bon résultats. Dans le chapitre 3, nous prouvons des résultats d'existence et d'unicité pour une classe générale d'équations progressives-rétrogrades à champs moyen sous une condition de monotonicité faible et une hypothèse non-dégénérescence sur l'équation progressive et nous donnons une application dans le domaine de stockage d'énergie dans le cas où la production d'électricité est imprévisible.

In this work, we investigate an optimal liquidation problem under Knightian uncertainty. We obtain the value function and an optimal control characterised by the solution of a second-order BSDE with monotone generator and with a singular terminal condition.

International audience; In the present paper, we aim at providing plug-in-type empirical estimators based on empirical processes to quantify the contribution of each operational or/and non-functioning state to the failures of a system described semi-Markov model. In particular we focus on an important reliability measure for random repairable systems, the rate of occurrence of failures and study different conditional versions in a discrete time and finite state space context. The aforementioned esti-mators of the conditional failure occurrence rates are caracterised by appealing asymptotic properties such as consistency and asymptotic normality. We further provide detailed analytical expressions for the covariance matrices of the random vectors describing the conditional failure occurrence rates. These results are illustrated on an academ-ical numerical example based on simulated data. We further present an application to a real dataset that models earthquake occurrences in a semi-Markov framework.

A growing number of studies indicate a widening of socioeconomic inequalities in mortality over the past decades. It has therefore become crucially important to understand the impact of heterogeneity and its evolution on the future mortality of heterogeneous populations. In particular, recent developments in multi-population mortality have raised a number of questions, among which is the issue of evaluating cause-of-death reduction targets set by national and international institutions in the presence of heterogeneity. The aim of this paper is to show how the study of the population data and the population dynamics framework contribute to addressing these issues, by providing a new viewpoint on the evolution of aggregate mortality indicators in the presence of heterogeneity. Our findings rely on two datasets on the English population and cause-specific number of deaths by socioeconomic circumstances, over the period 1981-2015. The analysis of the data first highlights the complexity of recent demographic developments, characterized by significant composition changes in the population, with considerable variations according to the age class or cohort, along with a widening of socioeconomic inequalities. We then introduce a dynamic framework for studying the impact of composition changes on the mortality of the global population. In particular, we are interested in quantifying the impacts of cause-of-death mortality reduction in comparison with changes of composition in a heterogeneous population. We show how a cause of death reduction could be compensated for in the presence of heterogeneity, which could lead to misinterpretations when assessing public policies impacts and/or for the forecasting of future trends.

We use the functional It{\^o} calculus to prove that the solution of a BSDE with singular terminal condition is continuous at the terminal time. Hence we extend known results for a non-Markovian terminal condition., Comment: arXiv admin note: substantial text overlap with arXiv:1601.03186

In this article, we are interested in solving numerically backward doubly stochastic differential equations (BDSDEs) with random terminal time tau. The main motivations are giving a probabilistic representation of the Sobolev's solution of Dirichlet problem for semilinear SPDEs and providing the numerical scheme for such SPDEs. Thus, we study the strong approximation of this class of BDSDEs when tau is the first exit time of a forward SDE from a cylindrical domain. Euler schemes and bounds for the discrete-time approximation error are provided., 38, Monte Carlo Methods and Applications (MCMA) 2016

International audience; An acoustic black hole (ABH) is a pit of power law profile inserted in a plate with internal damping. Such a pit with varying thickness leads to local variations of wave propagation properties and has been found to be an efficient vibration damper. In this paper, the ABH is seen as a penetrable obstacle and is studied through its scattering properties. A wave based model is developed within the framework of the Kirchhoff theory for a thin plate of locally varying thickness. It is shown that analytical solutions can be found in the case of a pit of quadratic profile and for uniform damping properties (without added layer). A major outcome of the model is the derivation of the dispersion relations giving a detailed analysis of the behavior of the waves within the ABH. Particularly, cut-on frequencies below which no wave propagates within the ABH are derived and thus give interpretation of the well known typical ABH efficiency threshold frequency for damping vibrations. The analysis is led from numerical computations of the dispersion curves, the scattering diagrams and the scattering cross-section, which are compared to the case of a simple hole. The results show that the ABH behaves as a resonant scatterer, which is a key outcome of this study. The so-called trapped modes, which describe free damped oscillations of the ABH, are responsible for variations of the scattering cross-section with frequency. These variations are investigated thanks to a parametric study on the geometrical properties of the ABH.

International audience; Mean times to failure are fundamental indicators in reliability and related fields. Here we focus on the conditional mean time to failure defined in a semi-Markov context. A discrete time semi-Markov model with discrete state space is employed, which allows for realistic description of systems under risk. Our main objective is to estimate the conditional mean time to failure and provide asymptotic properties of its nonparametric estimator. Consistency and asymptotic normality results are provided. Our methodology is tested in a real wind dataset and indicators associated with the wind energy production are estimated.

A new sequence of estimators is introduced to estimate the parameters of the fractional Gaussian noise at high-frequency. This sequence is defined by an initial sequence of quadratic generalized variations based estimators (QGV) and a single Fisher scoring step. It presents certain advantages over the sequence of maximum likelihood estima-tors (MLE) which is rate and variance efficient and the QGV which is only rate efficient. Indeed, it is much less computationally demanding than the MLE while keeping the efficient asymptotic variance. The local weak efficiency of crude oil markets is analyzed through likelihood-ratio hypothesis tests based on the joint estimation of the volatility and the Hurst exponent in the high-frequency scheme.

This paper is related to the study of systems of reflected backward stochastic differential equations with interconnected bilateral obstacles. These systems are connected with zero-sum stochastic switching games. Under appropriate assumptions, we provide either existence or existence and uniqueness of the solution of those systems when the switching costs are Itô processes. The link with systems of PDEs with bilateral interconnected obstacles is also stated via the Feynman-Kac representation when randomness comes from a Markov diffusion process. AMS Classification subjects: 93C30 ; 91A15 ; 93E20 ; 49J20.

In this article, we propose a wellposedness theory for a class of second order backward doubly stochastic differential equation (2BDSDE). We prove existence and uniqueness of the solution under a Lipschitz type assumption on the generator, and we investigate the links between our 2BDSDEs and a class of parabolic fully nonLinear Stochastic PDes. Precisely, we show that the Markovian solution of 2BDSDEs provide a probabilistic interpretation of the classical and stochastic viscosity solution of fully nonlinear SPDEs., 43

International audience; Generalized Linear Models with categorical explanatory variables are considered and parameters of the model are estimated with an original exact maximum likelihood method. The existence of a sequence of maximum likelihood estimators is discussed and considerations on possible link functions are proposed. A focus is then given on two particular positive distributions: the Pareto 1 distribution and the shifted log-normal distributions. Finally, the approach is illustrated on a actuarial dataset to model insurance losses.