Impact of Serial Correlation on Change Point Detection by Sparse Parameter Estimation
Autoři | |
---|---|
Rok publikování | 2011 |
Druh | Článek ve sborníku |
Konference | 10th International Conference APLIMAT 2011 |
Fakulta / Pracoviště MU | |
Citace | |
www | http://archiv.aplimat.com/2011/Proceedings/Modeling_and_Simulation/Neubauer_Vesely.pdf |
Obor | Aplikovaná statistika, operační výzkum |
Klíčová slova | change point detection; overparametrized model; sparse parameter estimation; serial correlation |
Popis | The contribution deals with change point detection in a one-dimensional stochastic process when using 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 applied to get sparse parameter estimates. Impact of serial correlations on change point detection is studied by simulations. |
Související projekty: |