Inference for Multiple Change Points in Time Series via Likelihood Ratio Scan Statistics
- 19 November 2015
- journal article
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series B: Statistical Methodology
- Vol. 78 (4) , 895-916
Abstract
Summary: We propose a likelihood ratio scan method for estimating multiple change points in piecewise stationary processes. Using scan statistics reduces the computationally infeasible global multiple-change-point estimation problem to a number of single-change-point detection problems in various local windows. The computation can be efficiently performed with order O{npt log (n)}. Consistency for the estimated numbers and locations of the change points are established. Moreover, a procedure is developed for constructing confidence intervals for each of the change points. Simulation experiments and real data analysis are conducted to illustrate the efficiency of the likelihood ratio scan method.Keywords
Funding Information
- HKSAR-RGC-ECS (405012)
- HKSAR-RGC-GRF (405113, 14601015)
This publication has 26 references indexed in Scilit:
- Multiscale Change Point InferenceJournal of the Royal Statistical Society Series B: Statistical Methodology, 2014
- Consistency of minimum description length model selection for piecewise stationary time series modelsElectronic Journal of Statistics, 2013
- Structural breaks in time seriesJournal of Time Series Analysis, 2012
- Testing for changes in the covariance structure of linear processesJournal of Statistical Planning and Inference, 2009
- Break Detection for a Class of Nonlinear Time Series ModelsJournal of Time Series Analysis, 2008
- Structural Break Estimation for Nonstationary Time Series ModelsJournal of the American Statistical Association, 2006
- Parametric Statistical Change Point Analysis2000
- Estimating Multiple Breaks One at a TimeEconometric Theory, 1997
- Mixing1994
- The Use of MOSUMS for Quality ControlTechnometrics, 1978