EcoSta 2018: Registration
View Submission - EcoSta2018
Title: Weighted adaptive hard threshold signal approximation Authors:  Xiaoli Gao - University of North Carolina at Greensboro (United States) [presenting]
Abstract: The aim is to formulate the copy number into a signal approximation model and to propose a robust change point detection method to simultaneously identify change points and outliers. This proposed method incorporates an individual weight for each observation and adopts the adaptive hard threshold approach to efficiently locate both outliers and copy number variations. The performance of the proposed robust signal approximation method is demonstrated by both simulations and real data analysis. Some theoretical results are also investigated.