Monte-Carlo Simulation Artefacts in Physical Modeling due to Pseudo-Random Generators
Abstract
A set of vector algorithms for random vectors generation is tested by the authors. Only the most reliable algorithms are selected for testing. The special system of original tests is developed with the main criteria of the independence of random vector components and the equability of N-dimensional hypercube filling. Some algorithms are discovered to be bad for physical modeling, and for others recommendations for their use for different physical problems are given.
| You currently do not have access to the full text article. |
|---|


