Learning the Conditional Independence Structure of Stationary Time Series: A Multitask Learning Approach

Abstract

We propose a method for inferring the conditional independence graph (CIG) of a high-dimensional Gaussian vector time series (discrete-time process) from a finite-length observation. By contrast to existing approaches, we do not rely on a parametric process model (such as, e.g., an autoregressive model) for the observed random process. Instead, we only require certain smoothness properties (in the Fourier domain) of the process. The proposed inference scheme works even for sample sizes much smaller than the number of scalar process components if the true underlying CIG is sufficiently sparse. A theoretical performance analysis provides conditions which guarantee that the probability of the proposed inference method to deliver a wrong CIG is below a prescribed value. These conditions imply lower bounds on the sample size such that the new method is consistent asymptotically. Some numerical experiments validate our theoretical performance analysis and demonstrate superior performance of our scheme compared to an existing (parametric) approach in case of model mismatch.

to be submitted to IEEE Trans. Sig. Proc

References in corpus (7)

• Consistency of the group Lasso and multiple kernel learning
• High-dimensional Ising model selection using ${\ell }_1$-regularized logistic regression
• Reduce and Boost: Recovering Arbitrary Sets of Jointly Sparse Vectors
• Operator norm consistent estimation of large-dimensional sparse covariance matrices
• On the Linear Convergence of the Alternating Direction Method of Multipliers
• Learning Networks of Stochastic Differential Equations
• Learning Mixed Graphical Models

Cited by in corpus (8)

• Large Spectral Density Matrix Estimation by Thresholding
• On Sparse High-Dimensional Graphical Model Learning For Dependent Time Series
• Graph Learning for Spatiotemporal Signals with Long- and Short-Term Characterization
• On the Sample Complexity of Graphical Model Selection for Non-Stationary Processes
• Learning Networked Exponential Families with Network Lasso
• Multi-task Learning for Financial Forecasting
• Graphical LASSO Based Model Selection for Time Series
• Learning conditional independence structure for high-dimensional uncorrelated vector processes