Sparse grid time-discontinuous Galerkin method with streamline diffusion for transport equations

Option pricing with a direct adaptive sparse grid approach

This work proposes and analyzes an anisotropic sparse grid stochastic collocation
\nmethod for solving partial differential equations with random coefficients and forcing terms (input
\ndata of the model). The method consists of a Galerkin approximation in the space variables and a
\ncollocation, in probability space, on sparse tensor product grids utilizing either Clenshaw–Curtis or
\nGaussian knots. Even in the presence of nonlinearities, the collocation approach leads to the solution
\nof uncoupled deterministic problems, just as in the Monte Carlo method. This work includes a priori
\nand a posteriori procedures to adapt the anisotropy of the sparse grids to each given problem.
\nThese procedures seem to be very effective for the problems under study. The proposed method
\ncombines the advantages of isotropic sparse collocation with those of anisotropic full tensor product
\ncollocation: the first approach is effective for problems depending on random variables which weigh
\napproximately equally in the solution, while the benefits of the latter approach become apparent when
\nsolving highly anisotropic problems depending on a relatively small number of random variables, as
\nin the case where input random variables are Karhunen–Lo`eve truncations of “smooth” random
\nfields. This work also provides a rigorous convergence analysis of the fully discrete problem and
\ndemonstrates (sub)exponential convergence in the asymptotic regime and algebraic convergence in
\nthe preasymptotic regime, with respect to the total number of collocation points. It also shows
\nthat the anisotropic approximation breaks the curse of dimensionality for a wide set of problems.
\nNumerical examples illustrate the theoretical results and are used to compare this approach with
\nseveral others, including the standard Monte Carlo. In particular, for moderately large-dimensional
\nproblems, the sparse grid approach with a properly chosen anisotropy seems to be very efficient and
\nsuperior to all examined methods.

Summary A sparse grid approach to orbit uncertainty propagation is presented. Efficient and accurate uncertainty propagation methods for nonlinear dynamic systems have been of enormous interest to space object tracking. Recent methods include those based on the time evolution of the probability density function, the statistical moments, the random samples, or a sum of Gaussian components. The idea of the sparse grid method for orbit uncertainty propagation is to represent the initial uncertainty by a sparse grid, propagate the sparse grid points individually through the nonlinear orbit dynamics, and compute the statistical moments from the propagated sparse grid points. The Smolyak rule is used to generate the sparse grid through linear combinations of low-level tensor products; the number of the resultant sparse grid points is polynomial in the dimension of the state. The sparse grid based uncertainty propagation may be considered a generalization of the Unscented Transformation (UT) of the well-known Unscented Kalman Filter but can achieve much higher accuracy levels than UT. Two examples of orbit propagation in different dynamic environments are used to show the efficacy of the sparse grid approach.

We present the parallelization of a sparse grid finite element discretization of the Black-Scholes equation, which is commonly used for option pricing. Sparse grids allow to handle higher dimensional options than classical approaches on full grids, and can be extended to a fully adaptive discretization method. We introduce the algorithmical structure of efficient algorithms operating on sparse grids, and demonstrate how they can be used to derive an efficient parallelization with OpenMP of the Black-Scholes solver. We show results on different commodity hardware systems based on multi-core architectures with up to 8 cores, and discuss the parallel performance using Intel and AMD CPUs.

The multi-element probabilistic collocation method (ME-PCM): Error analysis and applications

The stationary monochromatic radiative transfer equation (RTE) is a partial differential transport equation stated on a five-dimensional phase space, the Cartesian product of physical and angular domain. We solve the RTE with a Galerkin FEM in physical space and collocation in angle, corresponding to a discrete ordinates method (DOM). To reduce the complexity of the problem and to avoid the “curse of dimension”, we adapt the sparse grid combination technique to the solution space of the RTE and show that we obtain a sparse DOM which uses essentially only as many degrees of freedom as required for a purely spatial transport problem. For smooth solutions, the convergence rates deteriorate only by a logarithmic factor. We compare the sparse DOM to the standard full DOM and a sparse tensor product approach developed earlier with Galerkin FEM in physical space and a spectral method in angle. Numerical experiments confirm our findings.

A proof of convergence for the combination technique for the Laplace equation using tools of symbolic computation

The hydraulic conductivity of a confined aquifer is estimated using the quasi-linear geostatistical approach (QLGA), based on measurements of dependent quantities such as the hydraulic head or the arrival time of a tracer. This requires the solution of the steady-state groundwater flow and solute transport equations, which are coupled by Darcy's law. The standard Galerkin finite element method (FEM) for the flow equation combined with the streamline diffusion method (SDFEM) for the transport equation is widely used in the hydrogeologists' community. This work suggests to replace the first by the two-point flux cell-centered finite volume scheme (CCFV) and the latter by the Discontinuous Galerkin (DG) method. The convection-dominant case of solute (contaminant) transport in groundwater has always posed a special challenge to numerical schemes due to non-physical oscillations at steep fronts. The performance of the DG method is experimentally compared to the SDFEM approach with respect to numerical stability, accuracy and efficient solvability of the occurring linear systems. A novel method for the reduction of numerical under- and overshoots is presented as a very efficient alternative to local mesh refinement. The applicability and software-technical integration of the CCFV/DG combination into the large-scale inversion scheme mentioned above is realized. The high-resolution estimation of a synthetic hydraulic conductivity field in a 3-D real-world setting is simulated as a showcase on Linux high performance computing clusters. The setup in this showcase provides examples of realistic flow fields for which the solution of the convection-dominant transport problem by the streamline diffusion method fails.

This work presents reduced models for pricing basket options with the Black-Scholes and the Heston model. Basket options lead to multi-dimensional partial differential equations (PDEs) that quickly become computationally infeasible to discretize on full tensor grids. We therefore rely on sparse grid discretizations of the PDEs, which allow us to cope with the curse of dimensionality to some extent. We then derive reduced models with proper orthogonal decomposition. Our numerical results with the Black-Scholes model show that sufficiently accurate results are achieved while gaining speedups between 80 and 160 compared to the high-fidelity sparse grid model for 2-, 3-, and 4-asset options. For the Heston model, results are presented for a single-asset option that leads to a two-dimensional pricing problem, where we achieve significant speedups with our model reduction approach based on high-fidelity sparse grid models.

In this work we provide a convergence analysis for the quasi-optimal version of the sparse-grids stochastic collocation method we presented in a previous work: "On the optimal polynomial approximation of stochastic PDEs by Galerkin and collocation methods" (Beck et al., Math Models Methods Appl Sci 22(09), 2012). The construction of a sparse grid is recast into a knapsack problem: a profit is assigned to each hierarchical surplus and only the most profitable ones are added to the sparse grid. The convergence rate of the sparse grid approximation error with respect to the number of points in the grid is then shown to depend on weighted summability properties of the sequence of profits. This is a very general argument that can be applied to sparse grids built with any uni-variate family of points, both nested and non-nested. As an example, we apply such quasi-optimal sparse grids to the solution of a particular elliptic PDE with stochastic diffusion coefficients, namely the "inclusions problem": we detail the convergence estimates obtained in this case using polynomial interpolation on either nested (Clenshaw---Curtis) or non-nested (Gauss---Legendre) abscissas, verify their sharpness numerically, and compare the performance of the resulting quasi-optimal grids with a few alternative sparse-grid construction schemes recently proposed in the literature.

In simulation technology, computationally expensive objective functions are often replaced by cheap surrogates, which can be obtained by interpolation. Full grid interpolation methods suffer from the so-called curse of dimensionality, rendering them infeasible if the parameter domain of the function is higher-dimensional (four or more parameters). Sparse grids constitute a discretization method that drastically eases the curse, while the approximation quality deteriorates only insignificantly. However, conventional basis functions such as piecewise linear functions are not smooth (continuously differentiable). Hence, these basis functions are unsuitable for applications in which gradients are required. One example for such an application is gradient-based optimization, in which the availability of gradients greatly improves the speed of convergence and the accuracy of the results. This thesis demonstrates that hierarchical B-splines on sparse grids are well-suited for obtaining smooth interpolants for higher dimensionalities. The thesis is organized in two main parts: In the first part, we derive new B-spline bases on sparse grids and study their implications on theory and algorithms. In the second part, we consider three real-world applications in optimization: topology optimization, biomechanical continuum-mechanics, and dynamic portfolio choice models in finance. The results reveal that the optimization problems of these applications can be solved accurately and efficiently with hierarchical B-splines on sparse grids.

Sparse grids provide a very efficient method for the multilinear approximation of functions, especially in higher-dimensional spaces. In the d-dimensional space, the nodal multilinear basis on a grid with mesh size $h = 2^{ - n} $ consists of $O(2^{nd} )$ basis functions and leads to an $L\_2$-error of order $O(4^{ - n} )$ and an $H\_1$-error of order $O(2^{ - n} )$. -With sparse grids we get an $L\_2$-error of order $O(4^{ - n} n^{d - 1} )$ and an $H\_1$-error of order $O(2^{ - n} )$ with only $O(2^n n^{d - 1} )$ basis functions, if the function u fulfills the condition $\frac{{\partial ^{2d} }} {{\partial x_1^2 \partial x_2^2 \cdots \partial x_d^2 }}u < \infty $. Therefore, we can achieve much more accurate approximations with the same amount of storage. A data structure for the sparse grid representation of functions defined on cubes of arbitrary dimension and a finite element approach for the Helmholtz equation with sparse grid functions are introduced. Special emphasis is taken in the development of an efficient algorithm for the multiplication with the stiffness matrix. With an appropriate preconditioned conjugate gradient method (cg-method), the linear systems can be solved efficiently. Numerical experiments are presented for Helmholtz equations and eigenvalue problems for the Laplacian in two and three dimensions, and for a six-dimensional Poisson problem. The results support the assertion that the $L\_2$-error bounds for the sparse-grid approximation are also valid for sparse grid finite element solutions of elliptic differential equations. Problems with nonsmooth solutions are treated with adaptive sparse grids.

In this work we focus on the numerical approximation of the solution u of a linear elliptic PDE with stochastic coefficients. The problem is rewritten as a parametric PDE and the functional dependence of the solution on the parameters is approximated by multivariate polynomials. We first consider the stochastic Galerkin method, and rely on sharp estimates for the decay of the Fourier coefficients of the spectral expansion of u on an orthogonal polynomial basis to build a sequence of polynomial subspaces that features better convergence properties, in terms of error versus number of degrees of freedom, than standard choices such as Total Degree or Tensor Product subspaces. We consider then the Stochastic Collocation method, and use the previous estimates to introduce a new class of Sparse Grids, based on the idea of selecting a priori the most profitable hierarchical surpluses, that, again, features better convergence properties compared to standard Smolyak or tensor product grids. Numerical results show the effectiveness of the newly introduced polynomial spaces and sparse grids.

In this paper a new data structure for the distributed representation and efficient handling of sparse grids on parallel architectures is introduced. The new data layout is making use of the message passing paradigm and provides dynamic partioning and load balancing transparent to the application making use of it. This way sequential partial differential equations (PDE) solvers based on sparse grids can be data parallelized with hardly any source code modifications.

In this article, we propose the sparse grid combination technique for the second moment analysis of elliptic partial differential equations on random domains. By employing shape sensitivity analysis, we linearize the influence of the random domain perturbation on the solution. We derive deterministic partial differential equations to approximate the random solution’s mean and its covariance with leading order in the amplitude of the random domain perturbation. The partial differential equation for the covariance is a tensor product Dirichlet problem which can efficiently be determined by Galerkin’s method in the sparse tensor product space. We show that this Galerkin approximation coincides with the solution derived from the combination technique if the detail spaces in the related multiscale hierarchy are constructed with respect to Galerkin projections. This means that the combination technique does not impose an additional error in our construction. Numerical experiments quantify and qualify the proposed method.