A great deal of the statistical literature deals with a single sample coming from a
distribution, univariate or multivariate. As such this practice overlooks potentially
useful information from other sources including information from artificial sources.
An out of sample fusion method will be presented where an original real sample
is fused or combined with independent computer generated samples in the
estimation of very small tail probabilities under certain assumptions.  More generally,
out of sample fusion opens the door to a bootstrap counterpart where inference is
based on external rather than within sample data