Statistics Seminars: Multiscale MOSUM procedure with localised pruning
10 May 2018 16:00 in CM221
In this work, we investigate the problem of multiple change-point detection where the changes occur in the mean of univariate data. The proposed localised pruning methodology is applicable when conflicting change-point estimates are returned with the information about the local interval in which they are detected. We study the theoretical consistency of the localised pruning in combination with the multiscale extension of the MOving SUM (MOSUM) procedure Eichinger and Kirch (2018). Extensive simulation studies show the computational efficiency and good finite sample performance of the combined methodology, which is available as an R package 'mosum'.
This is joint work with Claudia Kirch (OvGU Magdeburg).
Contact firstname.lastname@example.org for more information