Event(s) on September 2021

• Thursday, 30th September, 2021

  Title:	Automatic Change Point Detection and Segment Estimation via Variational Bayesian Model Selection
  Speaker:	Mr DONG Qishi, Department of Mathematics, Hong Kong Baptist University, Hong Kong
  Time/Place:	16:00  -  17:30 Zoom, Meeting ID: 963 6974 9369 Password: 778799
  Abstract:	Change-point detection has long been an active research area, especially in the Big Data era, where data streams are usually non-stationary. However, challenges raised as many existing methods are only capable of single dimensional case, or require prior knowledge of number of change points, and usually need subsequent estimations for analysis. In this paper, we introduce an offline Bayesian change point model and associated scalable variational EM algorithm which can automatically estimate the number of change points and within characteristics in each segment. The comprehensive simulations for a normal mean-variance shift model and a discrete Poisson model and a real application in finance demonstrate the advantages of our approach. All the change-point locations and posterior inferences indicate the proposed method is comparable in location and parameter estimation, and computational efficiency over existing methods.

All years