A0686
Title: Multiple change point detection in functional sample via G-sum process
Authors: Tadas Danielius - Vilnius University (Lithuania) [presenting]
Alfredas Rackauskas - Vilnius University (Lithuania)
Abstract: The $G$-cumsum process is defined and investigated in its theoretical aspects including asymptotic behavior. By choosing different sets $G$ some tests for multiple change points detection in functional samples are proposed. The proposed testing procedures are applied to the real-world neurophysiological data and demonstrate how it can identify the existence of the multiple change points and localize them.
