ROBUST HYPOTHESIS TESTING FOR OBSERVED POISSON PROCESSES WITH UNCERTAIN RATE FUNCTIONS.

The problem of robust design is considered in the context of testing hypothesis about the rate function of an observed point process. Designs which are insensitive to uncertainty in the rate function under the alternative hypothesis are developed by applying a minimax formulation to a generalized measure of signal-to-noise ratio. Uncertainty in the rate is modeled by using general classes of rate measures generated by Choquet 2-alternating capacities, and solutions are characterized for this case by a Radon-Nikodym type derivative between such classes.