Volume 25, Issue 2

This paper proposes a new third-order discretization algorithm for multidimensional Itô stochastic differential equations driven by Brownian motions. The scheme is constructed by the Euler–Maruyama scheme with a stochastic weight given by polynomials of Brownian motions, which is simply implemented by a Monte Carlo method. The method of Watanabe distributions on Wiener space is effectively applied in the computation of the polynomial weight of Brownian motions. Numerical examples are shown to confirm the accuracy of the scheme.

General expressions for anisotropic particle diffusion Monte Carlo (PDMC) in a d -dimensional space are presented. The calculations of ground state energy of a helium atom for solving the many-body Schrödinger equation is carried out by the proposed method. The accuracy and stability of the results are discussed relative to other alternative methods, and our experimental results within the statistical errors agree with the quantum Monte Carlo methods. We also clarify the benefits of the proposed method by modeling the quantum probability density of a free particle in a plane (energy eigenfunctions). The proposed model represents a remarkable improvement in terms of performance, accuracy and computational time over standard MCMC method.

In this paper a random walk on arbitrary rectangles (2D) and parallelepipeds (3D) algorithm is developed for solving transient anisotropic drift-diffusion-reaction equations. The method is meshless, both in space and time. The approach is based on a rigorous representation of the first passage time and exit point distributions for arbitrary rectangles and parallelepipeds. The probabilistic representation is then transformed to a form convenient for stochastic simulation. The method can be used to calculate fluxes to any desired part of the boundary, from arbitrary sources. A global version of the method we call here as a stochastic expansion from cell to cell (SECC) algorithm for calculating the whole solution field is suggested. Application of this method to solve a system of transport equations for electrons and holes in a semicoductor is discussed. This system consists of the continuity equations for particle densities and a Poisson equation for electrostatic potential. To validate the method we have derived a series of exact solutions of the drift-diffusion-reaction problem in a three-dimensional layer presented in the last section in details.

The stationary distribution is the key of any queueing system; its determination is sufficient to infer the corresponding characteristics. This paper deals with the Er/M/1{\mathrm{Er}/M/1} queue. Bayesian inference is developed to estimate the system parameters, specially the root of the relative equation which allows the determination of the stationary distribution. A numerical study is performed with MCMC methods to support the results, and a comparison with another existing method in the literature (moments method) is provided.

This paper proposes a new approach to solving Ito stochastic differential equations. It is based on the well-known Monte Carlo methods for solving integral equations (Neumann–Ulam scheme, Markov chain Monte Carlo). The estimates of the solution for a wide class of equations do not have a bias, which distinguishes them from estimates based on difference approximations (Euler, Milstein methods, etc.).

Estimating hyper-volumes of convex and non-convex sets are of interest in a number of areas. In this article we develop further a simple geometric Monte Carlo method, known also as the sample-mean method, which transforms the domain to an equivalent hyper-sphere with the same volume. We first examine the performance of the method to compute the volumes of star-convex unit balls and show that it gives accurate estimates of their volumes. We then examine the use of this method for computing the volumes of nonstar-shaped domains. In particular, we develop two algorithms, which couple the sample-mean method with algebraic and geometric techniques, to generate and compute the volumes of low-dimensional stability domains in parameter space.

In this paper, we propose a Latin hypercube sampling (LHS) number generator in C language under Linux called getLHS in order to compare both methods LHS and refined descriptive sampling (RDS) method. It was highly tested by adequate statistical tests and compared statistically to the getRDS number generator. We noticed that getRDS has passed all tests better than the proposed getLHS generator. A simulation of M/G/1 queues is performed using getRDS to sample inputs from the RDS method and getLHS to sample inputs from the LHS method. The results obtained through simulation demonstrate that the RDS method produces more accurate point estimates of the true parameters than the LHS method. Moreover, the RDS method can significantly improve the performance of the studied queues compared to the well-known LHS method since its variance reduction factor is quite good in almost all cases. It is then proved that RDS is an improvement over LHS at least on queues.