Regularized estimation in sparse high-dimensional time series models

Many scientific and economic problems involve the analysis of high-dimensional time series datasets. However, theoretical studies in high-dimensional statistics to date rely primarily on the assumption of independent and identically distributed (i.i.d.) samples. In this work, we focus on stable Gaussian processes and investigate the theoretical properties of $\ell _1$-regularized estimates in two important statistical problems in the context of high-dimensional time series: (a) stochastic regression with serially correlated errors and (b) transition matrix estimation in vector autoregressive (VAR) models. We derive nonasymptotic upper bounds on the estimation errors of the regularized estimates and establish that consistent estimation under high-dimensional scaling is possible via $\ell_1$-regularization for a large class of stable processes under sparsity constraints. A key technical contribution of the work is to introduce a measure of stability for stationary processes using their spectral properties that provides insight into the effect of dependence on the accuracy of the regularized estimates. With this proposed stability measure, we establish some useful deviation bounds for dependent data, which can be used to study several important regularized estimates in a time series setting.

[1]  U. Grenander,et al.  Toeplitz Forms And Their Applications , 1958 .

[2]  S. Parter Extreme eigenvalues of Toeplitz forms and applications to elliptic difference equations , 1961 .

[3]  I. A. Ibragimov,et al.  On The Spectrum Of Stationary Gaussian Sequences Satisfying the Strong Mixing Condition I. Necessary Conditions , 1965 .

[4]  I. A. Ibragimov,et al.  On the Spectrum of Stationary Gaussian Sequences Satisfying the Strong Mixing Condition. II. Sufficient Conditions. Mixing Rate , 1970 .

[5]  C. Sims MACROECONOMICS AND REALITY , 1977 .

[6]  D. B. Preston Spectral Analysis and Time Series , 1983 .

[7]  Pravin Varaiya,et al.  Stochastic Systems: Estimation, Identification, and Adaptive Control , 1986 .

[8]  M. Pourahmadi,et al.  The mixing rate of a stationary multivariate process , 1993 .

[9]  James D. Hamilton Time Series Analysis , 1994 .

[10]  Martin Eichenbaum,et al.  Monetary Policy Shocks: What Have We Learned and to What End?" in The Handbook of Macroeconomics , 1999 .

[11]  Jianqing Fan,et al.  Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties , 2001 .

[12]  Jean Boivin,et al.  Measuring the Effects of Monetary Policy: A Factor-Augmented Vector Autoregressive (FAVAR) Approach , 2003 .

[13]  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.

[14]  R. C. Bradley Basic properties of strong mixing conditions. A survey and some open questions , 2005, math/0511078.

[15]  E. Liebscher Towards a Unified Approach for Proving Geometric Ergodicity and Mixing Properties of Nonlinear Autoregressive Processes , 2005 .

[16]  C. De Mol,et al.  Forecasting Using a Large Number of Predictors: Is Bayesian Regression a Valid Alternative to Principal Components? , 2006, SSRN Electronic Journal.

[17]  Bernard Bercu,et al.  Exponential inequalities for self-normalized martingales with applications , 2007, 0707.3715.

[18]  Helmut Ltkepohl,et al.  New Introduction to Multiple Time Series Analysis , 2007 .

[19]  M. Pesaran,et al.  Infinite Dimensional VARs and Factor Models , 2009, SSRN Electronic Journal.

[20]  S. Geer,et al.  On the conditions used to prove oracle results for the Lasso , 2009, 0910.0722.

[21]  Martin J. Wainwright,et al.  A unified framework for high-dimensional analysis of $M$-estimators with decomposable regularizers , 2009, NIPS.

[22]  Karl J. Friston Causal Modelling and Brain Connectivity in Functional Magnetic Resonance Imaging , 2009, PLoS biology.

[23]  P. Bickel,et al.  SIMULTANEOUS ANALYSIS OF LASSO AND DANTZIG SELECTOR , 2008, 0801.1095.

[24]  P. Bickel,et al.  Covariance regularization by thresholding , 2009, 0901.3079.

[25]  Cun-Hui Zhang Nearly unbiased variable selection under minimax concave penalty , 2010, 1002.4734.

[26]  Ali Shojaie,et al.  Discovering graphical Granger causality using the truncating lasso penalty , 2010, Bioinform..

[27]  Martin J. Wainwright,et al.  Estimation of (near) low-rank matrices with noise and high-dimensional scaling , 2009, ICML.

[28]  D. Giannone,et al.  Large Bayesian vector auto regressions , 2010 .

[29]  Shuheng Zhou Thresholded Lasso for high dimensional variable selection and statistical estimation , 2010, 1002.1583.

[30]  Martin J. Wainwright,et al.  Restricted Eigenvalue Properties for Correlated Gaussian Designs , 2010, J. Mach. Learn. Res..

[31]  W. Wu,et al.  Asymptotic theory for stationary processes , 2011 .

[32]  Lei Qi,et al.  Sparse High Dimensional Models in Economics. , 2011, Annual review of economics.

[33]  P. Bickel,et al.  Large Vector Auto Regressions , 2011, 1106.3915.

[34]  Po-Ling Loh,et al.  High-dimensional regression with noisy and missing data: Provable guarantees with non-convexity , 2011, NIPS.

[35]  Stephen M. Smith,et al.  The future of FMRI connectivity , 2012, NeuroImage.

[36]  Roman Vershynin,et al.  Introduction to the non-asymptotic analysis of random matrices , 2010, Compressed Sensing.

[37]  A. Kock,et al.  Oracle Inequalities for High Dimensional Vector Autoregressions , 2012, 1311.0811.

[38]  Han Xiao,et al.  Covariance matrix estimation for stationary time series , 2011, 1105.4563.

[39]  Richard A. Davis,et al.  Sparse Vector Autoregressive Modeling , 2012, 1207.0520.

[40]  Liudas Giraitis,et al.  Large Sample Inference for Long Memory Processes , 2012 .

[41]  Po-Ling Loh,et al.  Regularized M-estimators with nonconvexity: statistical and algorithmic theory for local optima , 2013, J. Mach. Learn. Res..

[42]  Yingying Fan,et al.  Asymptotic Equivalence of Regularization Methods in Thresholded Parameter Space , 2013, 1605.03310.

[43]  W. Wu,et al.  Covariance and precision matrix estimation for high-dimensional time series , 2013, 1401.0993.

[44]  Anil K. Seth,et al.  Granger causality analysis of fMRI BOLD signals is invariant to hemodynamic convolution but not downsampling , 2013, NeuroImage.

[45]  Fang Han,et al.  Transition Matrix Estimation in High Dimensional Time Series , 2013, ICML.

[46]  G. Michailidis,et al.  Autoregressive models for gene regulatory network inference: sparsity, stability and causality issues. , 2013, Mathematical biosciences.

[47]  M. Rudelson,et al.  Hanson-Wright inequality and sub-gaussian concentration , 2013 .

[48]  Jianqing Fan,et al.  Regularity Properties of High-dimensional Covariate Matrices ∗ , 2013 .

[49]  Shuheng Zhou,et al.  25th Annual Conference on Learning Theory Reconstruction from Anisotropic Random Measurements , 2022 .

[50]  Sumanta Basu,et al.  Modeling and Estimation of High-dimensional Vector Autoregressions. , 2014 .