AFFINITY IN PAIRED EVENT PROBABILITY

It is shown that a general parametric functional generates the conditi onal and joint probabilities of event pairs when the order within pair ed events is irrelevant. The parameters represent affinities or associ ations between single events. If the marginal probabilities of the sin gle events are known, then these parameters specify a hypersurface on which ail the joint probabilities of event pairs must lie. Examples ar e presented, and applications in probability, ecology, epidemiology, g enetics, and distribution theory are offered.