Estimation and inference for precision matrices of nonstationary time series

DOI10.1214/19-AOS1894zbMATH Open1471.62461arXiv1803.01188MaRDI QIDQ2215745FDOQ2215745

Authors: Xiucai Ding, Zhou Zhou 

Publication date: 14 December 2020

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: In this paper, we consider the estimation and inference of precision matrices of a rich class of locally stationary and nonlinear time series assuming that only one realization of the time series is observed. Using a Cholesky decomposition technique, we show that the precision matrices can be directly estimated via a series of least squares linear regressions with smoothly time-varying coefficients. The method of sieves is utilized for the estimation and is shown to be efficient and optimally adaptive in terms of estimation accuracy and computational complexity. We establish an asymptotic theory for a class of ${calL}^2$ tests based on the nonparametric sieve estimators. The latter are used for testing whether the precision matrices are diagonal or banded. A high dimensional Gaussian approximation result is established for a wide class of quadratic form of non-stationary and nonlinear processes, which is of interest by itself.

Full work available at URL: https://arxiv.org/abs/1803.01188

Recommendations

• Covariance and precision matrix estimation for high-dimensional time series
• Covariance matrix estimation for stationary time series
• Minimax estimation of covariance and precision matrices for high-dimensional time series with long-memory
• Estimation of large covariance and precision matrices from temporally dependent observations
• Local inference for locally stationary time series based on the empirical spectral measure

zbMATH Keywords

sieve estimationCholesky decompositionrandom matricesnonstationary time serieshigh-dimensional Gaussian approximationprecision matriceswhite noise and bandedness tests

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Nonparametric hypothesis testing (62G10) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Sparsistency and rates of convergence in large covariance matrix estimation
• Time series: theory and methods
• Gaussian approximation for high dimensional vector under physical dependence
• Regularized estimation of large covariance matrices
• Model selection and estimation in the Gaussian graphical model
• Decay Rates for Inverses of Band Matrices
• Nonlinear system theory: Another look at dependence
• Asymptotics of sample eigenstructure for a large dimensional spiked covariance model
• Rigidity of eigenvalues of generalized Wigner matrices
• Optimal global rates of convergence for nonparametric regression
• High dimensional inverse covariance matrix estimation via linear programming
• Joint mean-covariance models with applications to longitudinal data: unconstrained parameterisation
• Covariance and precision matrix estimation for high-dimensional time series
• Optimal uniform convergence rates and asymptotic normality for series estimators under weak dependence and weak conditions
• Nonparametric estimation of large covariance matrices of longitudinal data
• Covariance matrix estimation for stationary time series
• Some new asymptotic theory for least squares series: pointwise and uniform results
• An overview of the estimation of large covariance and precision matrices
• A general theory of concave regularization for high-dimensional sparse estimation problems
• Optimal rates of convergence for sparse covariance matrix estimation
• Local linear quantile estimation for nonstationary time series
• Optimal rates of convergence for estimating Toeplitz covariance matrices
• Spectral theory of block operator matrices and applications
• Random matrices: universality of local eigenvalue statistics
• Simultaneous Inference of Linear Models with Time Varying Coefficients
• Anisotropic local laws for random matrices
• A necessary and sufficient condition for edge universality at the largest singular values of covariance matrices
• Stein's method in high dimensions with applications
• An inversion algorithm for a banded matrix
• Inference for non-stationary time-series autoregression
• High-dimensional autocovariance matrices and optimal linear prediction

Cited In (12)

• Local Whittle estimation of high-dimensional long-run variance and precision matrices
• Estimation of high-dimensional dynamic conditional precision matrices with an application to forecast combination
• Graphical models for nonstationary time series
• Inference for high‐dimensional linear models with locally stationary error processes
• Random matrix models for datasets with fixed time horizons
• Multivariate functional response low‐rank regression with an application to brain imaging data
• Covariance and precision matrix estimation for high-dimensional time series
• StarTrek: combinatorial variable selection with false discovery rate control
• Estimation of banded time-varying precision matrix based on SCAD and group Lasso
• Inverse covariance operators of multivariate nonstationary time series
• Regularized Estimation of Linear Functionals of Precision Matrices for High-Dimensional Time Series
• Autoregressive approximations to nonstationary time series with inference and applications

This page was built for publication: Estimation and inference for precision matrices of nonstationary time series

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2215745)