COLLOQUIUM | Severien Nkurunziza (University of Windsor)
Oct 3, 2025
3:30PM to 4:30PM
Date/Time
Date(s) - 03/10/2025
3:30 pm - 4:30 pm
Location: Hamilton Hall, Room 305
Title: Change-points Modeling and Inference Methods In Some Mean-Reverting Processes
Speaker: Severien Nkurunziza (Professor, Mathematics and Statistics, University of Windsor)
Abstract: We provide an overview of recent developments in change-point modeling and inference related to the drift parameter in generalized mean-reverting processes. These processes are particularly well-suited for datasets characterized by a non-constant, periodically mean-reverting level, as well as by atypical or irregular shocks. We establish the unrestricted estimator (UE), the restricted estimator (RE), and a class of shrinkage estimators (SEs) for the drift parameter. In addition, we examine the relative efficiency of these estimators and derive an asymptotic test for hypotheses concerning the drift parameter. The proposed test is particularly useful for assessing the absence of a change-point, and we establish its consistency.
Coffee will be served in the same room, HH 305 at 3:00pm. All are welcome.