Testing for high-dimensional network parameters in auto-regressive models

DOI10.1214/19-EJS1646zbMATH Open1434.62201arXiv1812.03659MaRDI QIDQ2283570FDOQ2283570

Authors: L. L. Zheng, Garvesh Raskutti 

Publication date: 3 January 2020

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: High-dimensional auto-regressive models provide a natural way to model influence between $M$ actors given multi-variate time series data for $T$ time intervals. While there has been considerable work on network estimation, there is limited work in the context of inference and hypothesis testing. In particular, prior work on hypothesis testing in time series has been restricted to linear Gaussian auto-regressive models. From a practical perspective, it is important to determine suitable statistical tests for connections between actors that go beyond the Gaussian assumption. In the context of emph{high-dimensional} time series models, confidence intervals present additional estimators since most estimators such as the Lasso and Dantzig selectors are biased which has led to emph{de-biased} estimators. In this paper we address these challenges and provide convergence in distribution results and confidence intervals for the multi-variate AR(p) model with sub-Gaussian noise, a generalization of Gaussian noise that broadens applicability and presents numerous technical challenges. The main technical challenge lies in the fact that unlike Gaussian random vectors, for sub-Gaussian vectors zero correlation does not imply independence. The proof relies on using an intricate truncation argument to develop novel concentration bounds for quadratic forms of dependent sub-Gaussian random variables. Our convergence in distribution results hold provided $T=Omega((sveeho{)}^{2}lo{g}^{2}M)$, where $s$ and $ho$ refer to sparsity parameters which matches existed results for hypothesis testing with i.i.d. samples. We validate our theoretical results with simulation results for both block-structured and chain-structured networks.

Full work available at URL: https://arxiv.org/abs/1812.03659

Recommendations

• Bootstrap based inference for sparse high-dimensional time series models
• Gaussian approximation for high dimensional time series
• Adaptive test for mean vectors of high-dimensional time series data with factor structure
• Post-regularization inference for time-varying nonparanormal graphical models
• Hypothesis testing for high-dimensional time series via self-normalization

zbMATH Keywords

asymptotic normalityautoregressive modelssub-Gaussian distributionde-correlated score function

Mathematics Subject Classification ID

Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Estimation in multivariate analysis (62H12) Ridge regression; shrinkage estimators (Lasso) (62J07)

Cites Work

• Regularized estimation in sparse high-dimensional time series models
• Self-exciting point process modeling of crime
• Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
• Title not available (Why is that?)
• Confidence Intervals for Low Dimensional Parameters in High Dimensional Linear Models
• On asymptotically optimal confidence regions and tests for high-dimensional models
• A direct estimation of high dimensional stationary vector autoregressions
• Exact post-selection inference, with application to the Lasso
• Scaled sparse linear regression
• Network Estimation From Point Process Data
• Hanson-Wright inequality and sub-Gaussian concentration
• \(\ell_1\)-regularization of high-dimensional time-series models with non-Gaussian and heteroskedastic errors
• Combinatorics of partial derivatives
• Adaptive estimation of a quadratic functional by model selection.
• Autoregressive models for gene regulatory network inference: sparsity, stability and causality issues
• Testing for complete independence in high dimensions
• A general theory of hypothesis tests and confidence regions for sparse high dimensional models
• Tests for high-dimensional covariance matrices
• Graph estimation with joint additive models
• A test for the mean vector with fewer observations than the dimension under non-normality
• Structural changes in the cointegrated vector autoregressive model
• Toeplitz and circulant matrices: a review.
• Rejoinder: ``A significance test for the lasso
• An asymptotic expansion for probabilities of moderate deviations for multivariate martingales
• Variance Estimation Using Refitted Cross-Validation in Ultrahigh Dimensional Regression
• Tests for covariance matrix with fixed or divergent dimension
• A unified theory of confidence regions and testing for high-dimensional estimating equations
• The Componentwise Distance to the Nearest Singular Matrix
• Testing for trends in high-dimensional time series

Cited In (8)

• Does this suit me? Validation of self-modeling network models by parameter tuning
• Bootstrap based inference for sparse high-dimensional time series models
• High-Dimensional Vector Autoregressive Time Series Modeling via Tensor Decomposition
• Directed graphs and variable selection in large vector autoregressive models
• Rate-optimal robust estimation of high-dimensional vector autoregressive models
• Confidence intervals for parameters in high-dimensional sparse vector autoregression
• Structural inference in sparse high-dimensional vector autoregressions
• Statistical Inference for High-Dimensional Vector Autoregression with Measurement Error

Uses Software

• covTest

This page was built for publication: Testing for high-dimensional network parameters in auto-regressive models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2283570)