Adaptive quantile computation for Brownian bridge in change-point analysis

As an example for the fast calculation of distributional parameters Gaussian processes, a new Monte Carlo algorithm computation quantiles supremum norm weighted Brownian bridges is proposed. it known, corresponding distributions arise asymptotically CUSUM statistics change-point detection. The employs adaptive (sequential) time discretization trajectories bridge. A simulation study shows that by far outperforms standard approach, which uniform discretization.