A0748
Title: On detecting changes in certain random intensity-driven point processes through martingale-based repeated testing
Authors: Moinak Bhaduri - Bentley University (United States) [presenting]
Abstract: As surfacing, probably eventually, from the pandemic, other crises thwart normalcy: Appalling inequality, climate calamity, the banking crisis, upped possibilities of a fresh Cold War. The enduring motif of the time is incessant chaos. Frequently, much of that chaos results when one type of stationary system gives way to another. Change detection is mainly about estimating these points of deviation. In case a Poisson-type point process carries the system forward, a brand of online detection algorithms is offered, engineered through permutations of trend-switched statistics and a judicious application of false discovery rate control. Certain members of this family that remain asymptotically consistent and close to the ground truth (evidenced through some Hausdorff-similarity) are isolated to pinpoint estimated change locations. Efficient forecasting proves to be a natural corollary. Change point-based clustering tools will also be examined. it is described how such analyses offer concrete definitions to vague objects like COVID waves and measure their enormity.
