| Popis: |
Hitting time, or first passage time, is the first time that a process reaches (from above or below) a pre-specified area. Estimating expected hitting time is an important practical problem in many areas of science. In this paper, we discuss a novel and practical approach to estimating expected hitting time of [b, ∞), for a pre-specified boundary level b, as a function of the inverse of existing projections. An important example are predictions from an ensemble of models. Convergence properties of the estimator are established. Applications of the approach in epidemiology and climatology are presented. The performance of the approach is evaluated using real data from climate variability (El Niño) and flu prevalence. It is shown that the proposed approach has superior performance compared to commonly used methods in terms of MSE and other measures of goodness-of-fit. |
