Are deviations in a gradually varying mean relevant? A testing approach based on sup-norm estimators

Classical change point analysis aims at (1) detecting abrupt changes in the mean of a possibly non-stationary time series and at (2) identifying regions where the mean exhibits a piecewise constant behavior. In many applications however, it is more reasonable to assume that the mean changes gradually in a smooth way. Those gradual changes may either be non-relevant (i.e., small), or relevant for a specific problem at hand, and the present paper presents statistical methodology to detect the latter. More precisely, we consider the common nonparametric regression model $X_{i} = \mu (i/n) + \varepsilon_{i}$ with possibly non-stationary errors and propose a test for the null hypothesis that the maximum absolute deviation of the regression function $\mu$ from a functional $g (\mu )$ (such as the value $\mu (0)$ or the integral $\int_{0}^{1} \mu (t) dt$) is smaller than a given threshold on a given interval $[x_{0},x_{1}] \subseteq [0,1]$. A test for this type of hypotheses is developed using an appropriate estimator, say $\hat d_{\infty, n}$, for the maximum deviation $ d_{\infty}= \sup_{t \in [x_{0},x_{1}]} |\mu (t) - g( \mu) |$. We derive the limiting distribution of an appropriately standardized version of $\hat d_{\infty,n}$, where the standardization depends on the Lebesgue measure of the set of extremal points of the function $\mu(\cdot)-g(\mu)$. A refined procedure based on an estimate of this set is developed and its consistency is proved. The results are illustrated by means of a simulation study and a data example.

[1]  H. Dette,et al.  Detecting gradual changes in locally stationary processes , 2013, 1310.4678.

[2]  J. Steinebach,et al.  Testing for Changes in Multivariate Dependent Observations with an Application to Temperature Changes , 1999 .

[3]  Charlie Obimbo,et al.  Trend analysis and change point techniques: a survey , 2016 .

[4]  S. Wellek Testing Statistical Hypotheses of Equivalence and Noninferiority , 2010 .

[5]  J. Aitchison Confidence‐Region Tests , 1964 .

[6]  Gordon J Johnston,et al.  Probabilities of maximal deviations for nonparametric regression function estimates , 1982 .

[7]  M. B. Priestley,et al.  A Test for Non‐Stationarity of Time‐Series , 1969 .

[8]  N. Plummer,et al.  Trends in annual frequencies of extreme temperature events in Australia , 2000 .

[9]  Pavel Čížek,et al.  Robust nonparametric regression: A review , 2019, WIREs Computational Statistics.

[10]  P. Fryzlewicz Tail-greedy bottom-up data decompositions and fast multiple change-point detection , 2018, The Annals of Statistics.

[11]  B. Minasny The Elements of Statistical Learning, Second Edition, Trevor Hastie, Robert Tishirani, Jerome Friedman. (2009), Springer Series in Statistics, ISBN 0172-7397, 745 pp , 2009 .

[12]  Piotr Fryzlewicz,et al.  Multiple‐change‐point detection for high dimensional time series via sparsified binary segmentation , 2015, 1611.08639.

[13]  B. Hansen UNIFORM CONVERGENCE RATES FOR KERNEL ESTIMATION WITH DEPENDENT DATA , 2008, Econometric Theory.

[14]  Jianqing Fan,et al.  Local polynomial modelling and its applications , 1994 .

[15]  J. Aston,et al.  Evaluating stationarity via change-point alternatives with applications to fMRI data , 2012, 1301.2894.

[16]  Jianqing Fan Local Polynomial Modelling and Its Applications: Monographs on Statistics and Applied Probability 66 , 1996 .

[17]  Ming Fan Some inequalities for p-variations of martingales , 2004 .

[18]  Wei Biao Wu,et al.  CONFIDENCE BANDS IN NONPARAMETRIC TIME SERIES REGRESSION , 2008, 0808.1010.

[19]  P. Fryzlewicz,et al.  Narrowest‐over‐threshold detection of multiple change points and change‐point‐like features , 2016, Journal of the Royal Statistical Society: Series B (Statistical Methodology).

[20]  H. Ombao,et al.  Detection of Changes in Multivariate Time Series With Application to EEG Data , 2015 .

[21]  N. Vayatis,et al.  Selective review of offline change point detection methods , 2019 .

[22]  Fugee Tsung,et al.  Monitoring Profiles Based on Nonparametric Regression Methods , 2008, Technometrics.

[23]  Zhou Zhou,et al.  Heteroscedasticity and Autocorrelation Robust Structural Change Detection , 2013 .

[24]  Wei Biao Wu,et al.  Inference of trends in time series , 2007 .

[25]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[26]  Douglas A. Wolfe,et al.  Nonparametric statistical procedures for the changepoint problem , 1984 .

[27]  Irène Gijbels,et al.  Nonparametric simultaneous testing for structural breaks , 2008 .

[28]  Weidong Liu,et al.  Simultaneous nonparametric regression analysis of sparse longitudinal data , 2018, Bernoulli.

[29]  Soumendu Sundar Mukherjee,et al.  Weak convergence and empirical processes , 2019 .

[30]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[31]  Robert A. Lordo Image Processing and Jump Regression Analysis , 2006, Technometrics.

[32]  H. Dette,et al.  Detecting relevant changes in the mean of nonstationary processes—A mass excess approach , 2019 .

[33]  H. Dette,et al.  Multiscale change point detection for dependent data , 2018, Scandinavian Journal of Statistics.

[34]  S. Fotopoulos,et al.  Inference for single and multiple change‐points in time series , 2013 .

[35]  Zuofeng Shang,et al.  Variance Change Point Detection Under a Smoothly-Changing Mean Trend with Application to Liver Procurement , 2018, Journal of the American Statistical Association.

[36]  Charles W. Champ,et al.  Exact results for shewhart control charts with supplementary runs rules , 1987 .

[37]  Alessandro Casini,et al.  Structural Breaks in Time Series , 2018, Oxford Research Encyclopedia of Economics and Finance.

[38]  Douglas C. Montgomery,et al.  Research Issues and Ideas in Statistical Process Control , 1999 .

[40]  A. Munk,et al.  Multiscale change point inference , 2013, 1301.7212.

[41]  M. A. Graham,et al.  Nonparametric (distribution-free) control charts: An updated overview and some results , 2019, Quality Engineering.

[42]  Weidong Liu,et al.  Simultaneous Confidence Bands in Nonlinear Regression Models with Nonstationarity , 2015 .

[43]  C. Loader CHANGE POINT ESTIMATION USING NONPARAMETRIC REGRESSION , 1996 .

[44]  Douglas C. Montgomery,et al.  Some Current Directions in the Theory and Application of Statistical Process Monitoring , 2014 .

[45]  Zhou Zhou,et al.  Simultaneous inference of linear models with time varying coefficients , 2010 .

[46]  Thomas R. Karl,et al.  Trends in high-frequency climate variability in the twentieth century , 1995, Nature.

[47]  Irène Gijbels,et al.  Bandwidth Selection for Changepoint Estimation in Nonparametric Regression , 2004, Technometrics.

[48]  Peihua Qiu,et al.  Jump-Preserving Regression and Smoothing using Local Linear Fitting: A Compromise , 2007 .

[49]  W. Wu,et al.  Nonlinear system theory: another look at dependence. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[50]  Vladimir I. Piterbarg,et al.  Asymptotic Methods in the Theory of Gaussian Processes and Fields , 1995 .

[51]  Zhou Zhou,et al.  Local linear quantile estimation for nonstationary time series , 2009, 0908.3576.

[52]  Zhou Zhou,et al.  NONPARAMETRIC INFERENCE OF QUANTILE CURVES FOR NONSTATIONARY TIME SERIES , 2010, 1010.3891.

[53]  D. Hawkins,et al.  NONPARAMETRIC CONTROL CHART FOR MONITORING PROFILES USING CHANGE POINT FORMULATION AND ADAPTIVE SMOOTHING , 2009 .

[54]  H. Müller CHANGE-POINTS IN NONPARAMETRIC REGRESSION ANALYSIS' , 1992 .

[55]  K. Proksch On confidence bands for multivariate nonparametric regression , 2016 .

[56]  Zhijie Xiao,et al.  Testing structural change in time-series nonparametric regression models , 2008 .

[57]  Wei Biao Wu,et al.  STRONG INVARIANCE PRINCIPLES FOR DEPENDENT RANDOM VARIABLES , 2007, 0711.3674.

[58]  C. Chu,et al.  Kernel-Type Estimators of Jump Points and Values of a Regression Function , 1993 .

[59]  Manfred Mudelsee,et al.  Ramp function regression: a tool for quantifying climate transitions , 2000 .

[60]  W. R. Schucany,et al.  Improvement of Kernel Type Density Estimators , 1977 .

[61]  Yingcun Xia,et al.  Bias‐corrected confidence bands in nonparametric regression , 1998 .