Structural Break Estimation for Nonstationary Time Series Models

This article considers the problem of modeling a class of nonstationary time series using piecewise autoregressive (AR) processes. The number and locations of the piecewise AR segments, as well as the orders of the respective AR processes, are assumed unknown. The minimum description length principle is applied to compare various segmented AR fits to the data. The goal is to find the “best” combination of the number of segments, the lengths of the segments, and the orders of the piecewise AR processes. Such a “best” combination is implicitly defined as the optimizer of an objective function, and a genetic algorithm is implemented to solve this difficult optimization problem. Numerical results from simulation experiments and real data analyses show that the procedure has excellent empirical properties. The segmentation of multivariate time series is also considered. Assuming that the true underlying model is a segmented autoregression, this procedure is shown to be consistent for estimating the location of the breaks.

[1]  W. Gersch.  Spectral analysis of EEG's by autoregressive decomposition of time series , 1970 .

[2]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[3]  Genshiro Kitagawa,et al.  A procedure for the modeling of non-stationary time series , 1978 .

[4]  A. Hasman,et al.  Usefulness of autoregressive models to classify EEG segments , 2009 .

[5]  Yi-Ching Yao.  Estimating the number of change-points via Schwarz' criterion , 1988 .

[6]  Lawrence. Davis,et al.  Handbook Of Genetic Algorithms , 1990 .

[7]  Emergent computation , 1991 .

[8]  Richard A. Davis,et al.  Time Series: Theory and Methods , 2013 .

[9]  Richard A. Davis,et al.  Time Series: Theory and Methods (2nd ed.). , 1992 .

[10]  R. Blacher.  Higher Order Correlation Coefficients , 1993 .

[11]  G. C. Tiao,et al.  Use of Cumulative Sums of Squares for Retrospective Detection of Changes of Variance , 1994 .

[12]  Naoki Saito,et al.  Simultaneous noise suppression and signal compression using a library of orthonormal bases and the minimum-description-length criterion , 1994, Defense, Security, and Sensing.

[13]  P. Perron,et al.  Estimating and testing linear models with multiple structural changes , 1995 .

[14]  Richard A. Davis,et al.  Introduction to time series and forecasting , 1998 .

[15]  S. Adak.  Time dependent spectral analysis of nonstationary time series , 1996 .

[16]  Chung-Bow Lee,et al.  Estimating the Number of Change Points in Exponential Families Distributions , 1997 .

[17]  R. Dahlhaus.  Fitting time series models to nonstationary processes , 1997 .

[18]  Arjun K. Gupta,et al.  Testing and Locating Variance Changepoints with Application to Stock Prices , 1997 .

[19]  James P. Cohoon,et al.  C6.3 Island (migration) models: evolutionary algorithms based on punctuated equilibria , 1997 .

[20]  Jorma Rissanen,et al.  Stochastic Complexity in Statistical Inquiry , 1989, World Scientific Series in Computer Science.

[21]  P. Perron,et al.  Computation and Analysis of Multiple Structural-Change Models , 1998 .

[22]  Marc Lavielle,et al.  Optimal segmentation of random processes , 1998, IEEE Trans. Signal Process..

[23]  Chang‐Jin Kim,et al.  State Space Models with Regime Switching , 1999 .

[24]  Enrique Alba,et al.  A survey of parallel distributed genetic algorithms , 1999, Complex..

[25]  Bin Yu,et al.  Wavelet thresholding via MDL for natural images , 2000, IEEE Trans. Inf. Theory.

[26]  Thomas C.M. Lee.  A Minimum Description Length-Based Image Segmentation Procedure, and its Comparison with a Cross-Validation-Based Segmentation Procedure , 2000 .

[27]  Carlo Gaetan,et al.  Subset ARMA Model Identification Using Genetic Algorithms , 2000 .

[28]  Jennifer Pittman,et al.  Adaptive Splines and Genetic Algorithms , 2000 .

[29]  J. Raz,et al.  Automatic Statistical Analysis of Bivariate Nonstationary Time Series , 2001 .

[30]  Bin Yu,et al.  Model Selection and the Principle of Minimum Description Length , 2001 .

[31]  Christophe Andrieu,et al.  Bayesian curve fitting using MCMC with applications to signal segmentation , 2002, IEEE Trans. Signal Process..

[32]  Enrique Alba,et al.  Improving flexibility and efficiency by adding parallelism to genetic algorithms , 2002, Stat. Comput..

[33]  Dmitriy Melkonian,et al.  High-resolution fragmentary decomposition—a model-based method of non-stationary electrophysiological signal analysis , 2003, Journal of Neuroscience Methods.

[34]  Thomas C. M. Lee,et al.  Nonparametric log spectrum estimation using disconnected regression splines and genetic algorithms , 2003, Signal Process..

[35]  Bin Yu,et al.  Simultaneous Gene Clustering and Subset Selection for Sample Classification Via MDL , 2003, Bioinform..

[36]  S. Koopman,et al.  Extracting Business Cycles Using Semi-Parametric Time-Varying Spectra with Applications to Us Macroeconomic Time Series , 2006 .

[37]  Piotr Fryzlewicz,et al.  Haar–Fisz estimation of evolutionary wavelet spectra , 2006 .