Change Point Detection by Sparse Parameter Estimation

Article type: Research Article

Authors: Neubauer, Jiří | Veselý, Vítězslav

Affiliations: University of Defence, Kounicova 65, 612 00 Brno, Czech Republic, E-mail: jiri.neubauer@unob.cz | Masaryk University, Lipová 41a, 602 00 Brno, Czech Republic, E-mail: vesely@econ.muni.cz

Abstract: The contribution is focused on change point detection in a one-dimensional stochastic process by sparse parameter estimation from an overparametrized model. A stochastic process with change in the mean is estimated using dictionary consisting of Heaviside functions. The basis pursuit algorithm is used to get sparse parameter estimates. The mentioned method of change point detection in a stochastic process is compared with several standard statistical methods by simulations.

Keywords: change point detection, overparametrized model, sparse parameter estimation

Journal: Informatica, vol. 22, no. 1, pp. 149-164, 2011