Change point detection in a stochastic complexity framework
- 1 January 1990
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3554-3555 vol.6
- https://doi.org/10.1109/cdc.1990.203485
Abstract
The authors present a method, inspired by stochastic complexity theory, for solving the change point detection problem for ARMA (autoregressive moving average) systems which are assumed to have a slow unstructured nondecaying drift after the change has occurred. The central idea is to apply the minimum description length method in the form of predictive stochastic complexity, which gives a way for selecting the best model among a given set of models. Therefore the change point detection problem is reduced to a model selection problem. Simulations that show that the approach exhibits good detection capabilities are included.Keywords
This publication has 7 references indexed in Scilit:
- The two-models approach for the on-line detection of changes in AR processesPublished by Springer Nature ,2005
- Detection of abrupt changes in dynamic systemsPublished by Springer Nature ,2005
- On a class of mixing processesStochastics and Stochastic Reports, 1989
- Detecting changes in signals and systems—A surveyAutomatica, 1988
- Short-term load forecastingProceedings of the IEEE, 1987
- Stochastic Complexity and ModelingThe Annals of Statistics, 1986
- Uniqueness of prediction error estimates of multivariable moving average modelsAutomatica, 1982