Affordable Access

Publisher Website

Testing independence of simulation subruns: a note on the power of the Von Neumann test

Authors
Journal
European Journal of Operational Research
0377-2217
Publisher
Elsevier
Publication Date
Volume
9
Issue
1
Identifiers
DOI: 10.1016/0377-2217(82)90017-0

Abstract

Abstract In simulation experiments confidence intervals may be based on subruns. To test whether subruns are independent the Von Neumann statistic is often used. This note shows that the power of this statistic may be very small. The practical recommendation is to apply the statistic only if at least 100 subruns are available.

There are no comments yet on this publication. Be the first to share your thoughts.