1/(B+1) is the significance level of the Monte Carlo test if the data
give the most extreme value. (E.g. Ripley, 1987, p. 171.) This is a
calculation, not a convention, and assumes a continuously distributed
statistic.
On Mon, 8 Mar 2004, Spencer Graves wrote:
What is the standard
What is the standard convention for a Monte Carlo p-value when the
observed outcome is more extreme than all simulations? The example
provided by Cédric Fine produced a Monte Carlo p-value, according to
Kjetil Halvorsen, of 2.2e-16 (based on 2000 replicates). This seems
inappropriate to