Time Series Simulation with Quasi Monte Carlo Methods

Authors:
Jenny X. Li;Peter Winker
Affiliations:
Departments of Mathematics and Economics, The Pennsylvania State University, University Park, PA, 16802, U.S.A. li@math.psu.edu/;Department of Economics, University of Mannheim, D-68131 Mannheim, Germany Peter.Winker@vwl.uni-mannheim.de
Venue:
Computational Economics
Year:
2003

Citing 11
Cited 1

Stochastic simulation

Stochastic simulation
Random number generation and quasi-Monte Carlo methods

Random number generation and quasi-Monte Carlo methods
Numerical recipes in FORTRAN (2nd ed.): the art of scientific computing

Numerical recipes in FORTRAN (2nd ed.): the art of scientific computing
On the numerical integration of Walsh series by number-theoretic methods

Mathematics of Computation
Algorithm 659: Implementing Sobol's quasirandom sequence generator

ACM Transactions on Mathematical Software (TOMS)
Monte Carlo Variance of Scrambled Net Quadrature

SIAM Journal on Numerical Analysis
Application of Threshold-Accepting to the Evaluation of the Discrepancy of a Set of Points

SIAM Journal on Numerical Analysis
A generalized discrepancy and quadrature error bound

Mathematics of Computation
Uniform random number generators

Proceedings of the 30th conference on Winter simulation
Parallel computing of a quasi-Monte Carlo algorithm for valuing derivatives

Parallel Computing - Special issue on parallel computing in economics, finance and decision-making
Centered L2-discrepancy of random sampling and Latin hypercube design, and construction of uniform designs

Mathematics of Computation

Stochastic Ceteris Paribus Simulations

Computational Economics

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper compares quasi Monte Carlo methods, in particularso-called (t, m, s)-nets, with classical Monte Carlo approaches forsimulating econometric time-series models. Quasi Monte Carlomethods have found successful application in many fields, such asphysics, image processing, and the evaluation of financederivatives. However, they are rarely used in econometrics. Here,we apply both traditional and quasi Monte Carlo simulation methodsto time-series models that typically arise in macroeconometrics.The numerical experiments demonstrate that quasi Monte Carlomethods outperform traditional ones for all models we investigate.