Accelerated Monte Carlo estimation of exceedance probabilities under monotonicity constraints (Q1931811): Difference between revisions

The paper considers the problem of estimating \(p = P(g(X) \leq 0)\), the probability that a monotonic function \(g\) of a uniform \(d\)-dimensional random vector is smaller than zero. For small probabilities \(p\) the usual Monte Carlo estimator typically is too time consuming. An alternative iterative maximum likelihood estimator constructed form sequential nested Bernoulli data is introduced and its convergence properties are studied in detail. The favorable convergence properties of this estimator are studied in a series of examples.