In this thesis, we develop partial differential equation (PDE) based numerical methods to solve certain optimal stochastic control problems in finance. The value of a stochastic control problem is normally identical to the viscosity solution of a Hamilton-Jacobi-Bellman (HJB) equation or an HJB variational inequality. The HJB equation corresponds to the case when the controls are bounded while the HJB variational inequality corresponds to the unbounded control case. As a result, the solution to the stochastic control problem can be computed by solving the corresponding HJB equation/variational inequality as long as the convergence to the viscosity solution is guaranteed. We develop a unified numerical scheme based on a semi-Lagrangian timestepping for solving both the bounded and unbounded stochastic control problems as well as the discrete cases where the controls are allowed only at discrete times. Our scheme has the following useful properties: it is unconditionally stable; it can be shown rigorously to converge to the viscosity solution; it can easily handle various stochastic models such as jump diffusion and regime-switching models; it avoids Policy type iterations at each mesh node at each timestep which is required by the standard implicit finite difference methods. In this thesis, we demonstrate the properties of our scheme by valuing natural gas storage facilities---a bounded stochastic control problem, and pricing variable annuities with guaranteed minimum withdrawal benefits (GMWBs)---an unbounded stochastic control problem. In particular, we use an impulse control formulation for the unbounded stochastic control problem and show that the impulse control formulation is more general than the singular control formulation previously used to price GMWB contracts.

In this thesis, we develop partial differential equation (PDE) based numerical methods to solve certain optimal stochastic control problems in finance. The value of a stochastic control problem is normally identical to the viscosity solution of a Hamilton-Jacobi-Bellman (HJB) equation or an HJB variational inequality. The HJB equation corresponds to the case when the controls are bounded while the HJB variational inequality corresponds to the unbounded control case. As a result, the solution to the stochastic control problem can be computed by solving the corresponding HJB equation/variational inequality as long as the convergence to the viscosity solution is guaranteed. We develop a unified numerical scheme based on a semi-Lagrangian timestepping for solving both the bounded and unbounded stochastic control problems as well as the discrete cases where the controls are allowed only at discrete times. Our scheme has the following useful properties: it is unconditionally stable; it can be shown rigorously to converge to the viscosity solution; it can easily handle various stochastic models such as jump diffusion and regime-switching models; it avoids Policy type iterations at each mesh node at each timestep which is required by the standard implicit finite difference methods. In this thesis, we demonstrate the properties of our scheme by valuing natural gas storage facilities---a bounded stochastic control problem, and pricing variable annuities with guaranteed minimum withdrawal benefits (GMWBs)---an unbounded stochastic control problem. In particular, we use an impulse control formulation for the unbounded stochastic control problem and show that the impulse control formulation is more general than the singular control formulation previously used to price GMWB contracts.

Fundação Carlos Chagas Filho de Amparo a Pesquisa do Estado do Rio de Janeiro / Em suma, esta tese propõe uma metodologia de acompanhamento de interfaces móveis que baseia-se no método dos conjuntos de nível aqui chamado de método das curvas de nível, uma denominação baseada nas aplicações em que as interfaces são representadas por curvas acoplado a uma implementação semi-Lagrangeana, para problemas em meios porosos. Embora esta técnica possa, em princípio, ser aplicada a qualquer problema físico que apresente uma interface móvel, nesta tese são focados escoamentos em meios porosos consolidados e saturados por um ou dois fluidos imiscíveis e incompressíveis. Adicionalmente, um método iterativo paralelizável para a resolução de sistemas de equações lineares definidos em redes, que podem ser reduzidos à forma das equações fundamentais de equilíbrio, é empregado na determinação dos campos de velocidade associados aos escoamentos em meios porosos. O cenário semi-Lagrangeano acoplado ao método das curvas de nível é comparado com a implementação utilizando o bem conhecido esquema up-wind. Um exaustivo estudo realizado revela a superioridade da metodologia proposta frente à concorrente utilizando o up-wind. Finalmente, o método das curvas de nível com implementação semi-Lagrangeana (método semi-Lagrangeano das curvas de nível), e o método iterativo para a determinação do campo de velocidades são aplicados no estudo de problemas transientes em meios porosos que apresentam instabilidades dos tipos Saffman-Taylor e Rayleigh-Taylor. Este estudo envolve uma análise de estabilidade linear, a introdução de diversas perturbações trigonométricas na interface e a sua evolução não-linear. / Briefy, this thesis proposes a method for capturing moving interfaces based on the level set method coupled to a Semi-Lagrangian implementation for problems in porous media. Although this method could, in principle, be applied to any physical problem with moving interfaces, we foccus, in this thesis, on flows inside a consolidated porous media saturated by one or two imiscible and incompressible fluids. Besides, a parallelizable iterative method for solving linear systems defined on a network that can be reduced to the fundamental equilibrium equations, is employed to determine the velocity field associated with the flow in a porous medium. The semi-Lagrangian scheme coupled with the level set method is compared with the well-known implementation with the up-wind scheme. An exhaustive study is performed and reveals the superiority of the proposed scheme in relation to the competing one using the up-wind method. Finally, the level set method with semi-Lagrangian implementation and the iterative method for determining the velocity field are applied to the study of transient problems in porous media which present Saffman-Taylor and Rayleigh-Taylor instabilities. This study involves the application of a linear stability analysis, the introduction of several trigonometric perturbations to the interface and its non-linear evolution.

Fundação Carlos Chagas Filho de Amparo a Pesquisa do Estado do Rio de Janeiro / Em suma, esta tese propõe uma metodologia de acompanhamento de interfaces móveis que baseia-se no método dos conjuntos de nível aqui chamado de método das curvas de nível, uma denominação baseada nas aplicações em que as interfaces são representadas por curvas acoplado a uma implementação semi-Lagrangeana, para problemas em meios porosos. Embora esta técnica possa, em princípio, ser aplicada a qualquer problema físico que apresente uma interface móvel, nesta tese são focados escoamentos em meios porosos consolidados e saturados por um ou dois fluidos imiscíveis e incompressíveis. Adicionalmente, um método iterativo paralelizável para a resolução de sistemas de equações lineares definidos em redes, que podem ser reduzidos à forma das equações fundamentais de equilíbrio, é empregado na determinação dos campos de velocidade associados aos escoamentos em meios porosos. O cenário semi-Lagrangeano acoplado ao método das curvas de nível é comparado com a implementação utilizando o bem conhecido esquema up-wind. Um exaustivo estudo realizado revela a superioridade da metodologia proposta frente à concorrente utilizando o up-wind. Finalmente, o método das curvas de nível com implementação semi-Lagrangeana (método semi-Lagrangeano das curvas de nível), e o método iterativo para a determinação do campo de velocidades são aplicados no estudo de problemas transientes em meios porosos que apresentam instabilidades dos tipos Saffman-Taylor e Rayleigh-Taylor. Este estudo envolve uma análise de estabilidade linear, a introdução de diversas perturbações trigonométricas na interface e a sua evolução não-linear. / Briefy, this thesis proposes a method for capturing moving interfaces based on the level set method coupled to a Semi-Lagrangian implementation for problems in porous media. Although this method could, in principle, be applied to any physical problem with moving interfaces, we foccus, in this thesis, on flows inside a consolidated porous media saturated by one or two imiscible and incompressible fluids. Besides, a parallelizable iterative method for solving linear systems defined on a network that can be reduced to the fundamental equilibrium equations, is employed to determine the velocity field associated with the flow in a porous medium. The semi-Lagrangian scheme coupled with the level set method is compared with the well-known implementation with the up-wind scheme. An exhaustive study is performed and reveals the superiority of the proposed scheme in relation to the competing one using the up-wind method. Finally, the level set method with semi-Lagrangian implementation and the iterative method for determining the velocity field are applied to the study of transient problems in porous media which present Saffman-Taylor and Rayleigh-Taylor instabilities. This study involves the application of a linear stability analysis, the introduction of several trigonometric perturbations to the interface and its non-linear evolution.