Title: Monitoring in one and two sample multivariate situations and some modifications
Authors: Marie Huskova - Charles University (Czech Republic) [presenting]
Zdenek Hlavka - Charles University (Czech Republic)
Simos Meintanis - University of Athens (Greece)
Abstract: One and two sample multivariate change point detection procedures are considered. The proposed method is a L2-type criteria based on empirical characteristic functions. The focus is on on-line procedures. Asymptotic properties are presented together with results of a simulation study. The new method is also applied on a real data-set from the financial sector over a time period. Possible extensions will be discussed.