Abstract:Proofs of all technical lemmas, propositions and theorems. We propose NonStGM, a general nonparametric graphical modeling framework, for studying dynamic associations among the components of a nonstationary multivariate time series. It builds on the framework of Gaussian graphical models (GGM) and stationary time series graphical models (StGM) and complements existing works on parametric graphical models based on change point vector autoregressions (VAR). Analogous to StGM, the proposed framework captures conditional noncorrelations (both intertemporal and contemporaneous) in the form of an undirected graph. In addition, to describe the more nuanced nonstationary relationships among the components of the time series, we introduce the new notion of conditional nonstationarity/stationarity and incorporate it within the graph. This can be used to search for small subnetworks that serve as the “source” of nonstationarity in a large system. We explicitly connect conditional noncorrelation and stationarity between and within components of the multivariate time series to zero and Toeplitz embeddings of an infinite-dimensional inverse covariance operator. In the Fourier domain, conditional stationarity and noncorrelation relationships in the inverse covariance operator are encoded with a specific sparsity structure of its integral kernel operator. We show that these sparsity patterns can be recovered from finite-length time series by nodewise regression of discrete Fourier transforms (DFT) across different Fourier frequencies. We demonstrate the feasibility of learning NonStGM structure from data using simulation studies.

Generating Different Gaussian Multivariate Processes With Identical Graphs

Generation of Two Correlated Stationary Gaussian Processes

Hausdorff Dimension of Set Generated by Exceptional Oscillations of a Class of N-parameter Gaussian Processes

Learning Gaussian Networks

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

On the Information-theoretic Limits of Graphical Model Selection for Gaussian Time Series

Continuously Indexed Graphical Models

On Sparse High-Dimensional Graphical Model Learning For Dependent Time Series

An Efficient Approach to Graphical Modelling of Time Series

Graphical models for nonstationary time series

Graph Learning from Gaussian and Stationary Graph Signals

Compressive Nonparametric Graphical Model Selection For Time Series

Stationary Graph Processes and Spectral Estimation

Graph Classification Gaussian Processes via Spectral Features

The Gaussian Graphical Model in Cross-Sectional and Time-Series Data

Structured Sequential Gaussian Graphical Models for Monitoring Time-Varying Process

Gaussian Whittle-Matérn fields on metric graphs

Graph Classification Gaussian Processes via Hodgelet Spectral Features

Spatiotemporal Hawkes processes with a graphon-induced connectivity structure

Stochastic Deep Gaussian Processes over Graphs