Speaker：Junjian Zhang (Guangxi Normal University)

Time：2023-10-27, 10:00

Location：Conference Room 686 at the 6th floor of Shuli Building at Haiyun Campus

Abstract:

Change-point problem is widely appeared in many situations such as signal processing, climate, network flow and so on. In this talk, the mixed regression model is proposed, which considers both the covariates and their lag term. Change-point estimation is investigated by using the penalized least square method, quantile regression method, linear expectile and adaptive LASSO method for the proposed model. The corresponding statistical properties are given. Simulations and real analysis are done to show the advantage of the proposed model and their estimations.