A simple approximation for the distribution of the weighted combination of non-independent or independent probabilities

This paper presents a simple approximation for the distribution of the weighted Fisher's statistic X = -2log (P1P2w2)-P-w1... P-k(wk), which is suitable for use when P-1,P-2,..., P-k are non-independent or independent tail-area probabilities arising from k continuous distributions, and w(1), w(2),..., w(k) are constants. (c) 2005 Published by Elsevier B.V.