Abstract

In this talk, we consider the inference problem concerning the drift parameter in the generalized mean-reverting process, which is suitable for modeling the data with periodic features. We also study the scenario where there may exist some linear restrictions on the drift parameters. We propose the unrestricted estimator, restricted estimator, and shrinkage estimator. Further, we assess the relative efficiency of different estimators. Additionally, we investigate the detection of change-point within this framework. Our simulation studies demonstrate the applicability of the proposed methods, and we apply the proposed method to real-world
environmental data. These findings underscore the importance of precise inference and timely detection of changes in environmental processes. Finally, the proposed method is expected to contribute to a better understanding and management of environmental systems.