The Annals of Statistics
- Ann. Statist.
- Volume 46, Number 3 (2018), 1197-1224.
Gradient-based structural change detection for nonstationary time series M-estimation
We consider structural change testing for a wide class of time series M-estimation with nonstationary predictors and errors. Flexible predictor-error relationships, including exogenous, state-heteroscedastic and autoregressive regressions and their mixtures, are allowed. New uniform Bahadur representations are established with nearly optimal approximation rates. A CUSUM-type test statistic based on the gradient vectors of the regression is considered. In this paper, a simple bootstrap method is proposed and is proved to be consistent for M-estimation structural change detection under both abrupt and smooth nonstationarity and temporal dependence. Our bootstrap procedure is shown to have certain asymptotically optimal properties in terms of accuracy and power. A public health time series dataset is used to illustrate our methodology, and asymmetry of structural changes in high and low quantiles is found.
Ann. Statist., Volume 46, Number 3 (2018), 1197-1224.
Received: August 2016
Revised: February 2017
First available in Project Euclid: 3 May 2018
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Zentralblatt MATH identifier
Wu, Weichi; Zhou, Zhou. Gradient-based structural change detection for nonstationary time series M-estimation. Ann. Statist. 46 (2018), no. 3, 1197--1224. doi:10.1214/17-AOS1582. https://projecteuclid.org/euclid.aos/1525313080
- Supplement to “Gradient-based structural change detection for nonstationary time series M-estimation”. We provide (a) detailed proofs of the theorems and lemmas, (b) theoretical investigation on parameter estimation under the alternative hypothesis, (c) the analysis of dynamic models, (d) extension of our methodology to finitely many M-estimations and (e) extra simulation results.