Numerical Experiments for Adaptive Wavelet Methods for Stochastic Poisson Equations

Stefan Kinzel, Universit├Ąt Marburg

Abstract: We are interested in adaptive numerical wavelet algorithms for elliptic boundary value problems with a random right-hand side. The Poisson equation on a Lipschitz domain serves as a model problem. The stochastic model for the right-hand side provides explicit control of the Besov regularity and, in particular, permits sparsity in a wavelet expansion.
We investigate the average error and cost of an adaptive approximation to the solution.

Supported by the DFG within the SPP 1324.