A Measure of Stationarity in Locally Stationary Processes With Applications to Testing

In this article we investigate the problem of measuring deviations from stationarity in locally stationary time series. Our approach is based on a direct estimate of the L2-distance between the spectral density of the locally stationary process and its best approximation by a spectral density of a stationary process. An explicit expression of the minimal distance is derived, which depends only on integrals of the spectral density of the locally stationary process and its square. These integrals can be estimated directly without estimating the spectral density, and as a consequence, the estimation of the measure of stationarity does not require the specification of a smoothing bandwidth. We show weak convergence of an appropriately standardized version of the statistic to a standard normal distribution. The results are used to construct confidence intervals for the measure of stationarity and to develop a new test for the hypothesis of stationarity. Finally, we investigate the finite sample properties of the resulting confidence intervals and tests by means of a simulation study and illustrate the methodology in two data examples. Parts of the proofs are available online as supplemental material to this article.

[1]  R. Dahlhaus.  On the Kullback-Leibler information divergence of locally stationary processes , 1996 .

[2]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[3]  Rainer von Sachs,et al.  Locally adaptive estimation of evolutionary wavelet spectra , 2008, 0808.1452.

[4]  Masanobu Taniguchi,et al.  Discriminant analysis for locally stationary processes , 2004 .

[5]  Philip Preuss,et al.  Comparing spectral densities of stationary time series with unequal sample sizes , 2012 .

[6]  R. Dahlhaus,et al.  Nonparametric quasi-maximum likelihood estimation for Gaussian locally stationary processes , 2006, 0708.0143.

[7]  R. Dahlhaus.  Local inference for locally stationary time series based on the empirical spectral measure , 2009 .

[8]  K. Sakiyama,et al.  Testing Composite Hypotheses for Locally Stationary Processes , 2003 .

[9]  M. Priestley.  Evolutionary Spectra and Non‐Stationary Processes , 1965 .

[10]  SeriesSudeshna Adak.  Time-dependent Spectral Analysis ofNonstationary Time , 1995 .

[11]  Rainer Dahlhaus,et al.  A Likelihood Approximation for Locally Stationary Processes , 2000 .

[12]  A. T. C. Goh,et al.  Nonlinear modeling with confidence estimation using Bayesian neural networks , 2004 .

[13]  Rainer von Sachs,et al.  A Wavelet‐Based Test for Stationarity , 2000 .

[14]  Efstathios Paparoditis,et al.  Validating Stationarity Assumptions in Time Series Analysis by Rolling Local Periodograms , 2010 .

[15]  Okyoung Na,et al.  The Cusum Test for Parameter Change in Time Series Models , 2003 .

[16]  Paul Newbold,et al.  Testing the equality of prediction mean squared errors , 1997 .

[17]  R. Dahlhaus,et al.  Empirical spectral processes for locally stationary time series , 2009, 0902.1448.

[18]  E. Paparoditis,et al.  Frequency Domain Tests of Semiparametric Hypotheses for Locally Stationary Processes , 2009 .

[19]  J. Berger,et al.  Testing Precise Hypotheses , 1987 .

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

[21]  G. Nason,et al.  Wavelet processes and adaptive estimation of the evolutionary wavelet spectrum , 2000 .

[22]  Masanobu Taniguchi.  ON ESTIMATION OF THE INTEGRALS OF CERTAIN FUNCTIONS OF SPECTRAL DENSITY , 1980 .

[23]  Rainer von Sachs,et al.  Wavelet thresholding in anisotropic function classes and application to adaptive estimation of evolutionary spectra , 1997 .

[24]  D. Brillinger.  Time series - data analysis and theory , 1981, Classics in applied mathematics.

[25]  David J. C. MacKay,et al.  A Practical Bayesian Framework for Backpropagation Networks , 1992, Neural Computation.

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

[27]  Efstathios Paparoditis,et al.  Testing temporal constancy of the spectral structure of a time series , 2009, 1001.2122.