B1200
Title: A random projection based technique for change point estimation in high dimension
Authors: Nilabja Guha - UMASS Lowell (United States) [presenting]
Jyotishka Datta - Virginia Polytechnic Institute and State University (United States)
Abstract: A Bayesian framework of change point estimation for high-dimensional observations is presented. Such high-dimensional observations may appear in many practical applications where the high-dimensional mean parameter changes with time. A lower dimensional embedding is presented based on random projection. Change point estimation consistency is established, and convergence rate is established even when the dimension of the observations is much larger than the number of observations. Results are shown under known and unknown covariance structures, and related examples and applications are explored.
