Robust estimation of the scale and of the autocovariance function of Gaussian short- and long-range dependent processes

DOI10.1111/J.1467-9892.2010.00688.XzbMATH Open1290.62082arXiv0912.4686OpenAlexW2102680499MaRDI QIDQ4979097FDOQ4979097

Authors: Murad S. Taqqu, Valdério Anselmo Reisen, Céline Lévy-Leduc, Helene Boistard, Eric Moulines 

Publication date: 16 June 2014

Published in: Journal of Time Series Analysis (Search for Journal in Brave)

Abstract: A desirable property of an autocovariance estimator is to be robust to the presence of additive outliers. It is well-known that the sample autocovariance, being based on moments, does not have this property. Hence, the use of an autocovariance estimator which is robust to additive outliers can be very useful for time-series modeling. In this paper, the asymptotic properties of the robust scale and autocovariance estimators proposed by Rousseeuw and Croux (1993) and Genton and Ma (2000) are established for Gaussian processes, with either short-range or long-range dependence. It is shown in the short-range dependence setting that this robust estimator is asymptotically normal at the rate $sqrtn$, where $n$ is the number of observations. An explicit expression of the asymptotic variance is also given and compared to the asymptotic variance of the classical autocovariance estimator. In the long-range dependence setting, the limiting distribution displays the same behavior than that of the classical autocovariance estimator, with a Gaussian limit and rate $sqrtn$ when the Hurst parameter $H$ is less 3/4 and with a non-Gaussian limit (belonging to the second Wiener chaos) with rate depending on the Hurst parameter when $Hin(3/4,1)$. Some Monte-Carlo experiments are presented to illustrate our claims and the Nile River data is analyzed as an application. The theoretical results and the empirical evidence strongly suggest using the robust estimators as an alternative to estimate the dependence structure of Gaussian processes.

Full work available at URL: https://arxiv.org/abs/0912.4686

Recommendations

• Highly robust estimation of the autocovariance function
• Central limit theorem for the robust log-regression wavelet estimation of the memory parameter in the Gaussian semi-parametric context
• On a class of M-estimators for Gaussian long-memory models
• Gaussian semiparametric estimation of long range dependence
• Robust estimation in parametric time series models under long- and short-range-dependent structures

zbMATH Keywords

robustnessinfluence functionlong-memoryautocovariance functionHadamard differentiabilityscale estimatorfunctional Delta method

Mathematics Subject Classification ID

Nonparametric robustness (62G35) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Gaussian processes (60G15) Non-Markovian processes: estimation (62M09) Central limit and other weak theorems (60F05)

Cites Work

• Title not available (Why is that?)
• Highly robust estimation of the autocovariance function
• Robust estimation in long-memory processes under additive outliers
• Time series: theory and methods.
• Asymptotic Statistics
• Robust Statistics
• Fractional differencing
• Alternatives to the Median Absolute Deviation
• Robust Statistics
• Gaussian semiparametric estimation of long range dependence
• Title not available (Why is that?)
• Weak convergence to fractional brownian motion and to the rosenblatt process
• The empirical process of some long-range dependent sequences with an application to U-statistics
• A test for additive outliers applicable to long-memory time series
• Central limit theorems for non-linear functionals of Gaussian fields
• Limit theorems for functionals of mixing processes with applications to \(U\)-statistics and dimension estimation
• The empirical process of a short-range dependent stationary sequence under Gaussian subordination
• Limit theorems for nonlinear functionals of a stationary Gaussian sequence of vectors
• Multiple stochastic integrals with dependent integrators
• Bivariate symmetric statistics of long-range dependent observations
• Asymptotic properties of \(U\)-processes under long-range dependence

Cited In (22)

• Robust estimation of fractional seasonal processes: modeling and forecasting daily average \(\mathrm{SO}_2\) concentrations
• A dimension reduction factor approach for multivariate time series with long-memory: a robust alternative method
• Robust factor modelling for high-dimensional time series: an application to air pollution data
• An \(M\)-estimator for the long-memory parameter
• A theory of robust long-run variance estimation
• Robust factor models for high-dimensional time series and their forecasting
• The difference of symmetric quantiles under long range dependence
• A spectral approach to estimate the autocovariance function
• \(M\)-periodogram for the analysis of long-range-dependent time series
• On model Fitting and estimation of strictly stationary processes
• Highly robust estimation of the autocovariance function
• Large sample behaviour of some well-known robust estimators under long-range dependence
• Asymptotic properties of \(U\)-processes under long-range dependence
• Central limit theorem for the robust log-regression wavelet estimation of the memory parameter in the Gaussian semi-parametric context
• Robust estimation of periodic autoregressive processes in the presence of additive outliers
• Bayesian inference for ARFIMA models
• Fractional Brownian motion: difference iterative forecasting models
• A breakpoint detection in the mean model with heterogeneous variance on fixed time intervals
• Robust Two-Step Wavelet-Based Inference for Time Series Models
• Robust Dickey-Fuller tests based on ranks for time series with additive outliers
• Title not available (Why is that?)
• How the instability of ranks under long memory affects large-sample inference

Uses Software

• longmemo

This page was built for publication: Robust estimation of the scale and of the autocovariance function of Gaussian short- and long-range dependent processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4979097)