Title: Change point detection in VAR models based on characteristic functions
Authors: Zdenek Hlavka - Charles University (Czech Republic) [presenting]
Marie Huskova - Charles University (Czech Republic)
Simos Meintanis - University of Athens (Greece)
Abstract: A new method for detecting structural breaks in multivariate time series (VAR models) is proposed using L2-type criteria based on the empirical characteristic function (CF). The advantage of using CF-based procedure is that vector observations are linearly projected onto the real line and the resulting statistic may be written in a convenient closed-form expression. Asymptotic as well as Monte-Carlo results are presented. The new method is applied to time-series data from the financial sector.