Dynamic networks with multi-scale temporal structure

DOI10.1007/S13171-021-00256-1MaRDI QIDQ2121707zbMATH OpenOpenAlexFDO

Authors Xinyu Kang, Apratim Ganguly, Eric D. Kolaczyk

Publication date 4 April 2022

Published in Sankhyā. Series A (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1712.08586

zbMATH Keywords

vector autoregressive modeldynamic networkmultiscale modeling

Mathematics Subject Classification ID

Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Applications of statistics to biology and medical sciences; meta analysis (62P10) Small world graphs, complex networks (graph-theoretic aspects) (05C82)

Abstract: We describe a novel method for modeling non-stationary multivariate time series, with time-varying conditional dependencies represented through dynamic networks. Our proposed approach combines traditional multi-scale modeling and network based neighborhood selection, aiming at capturing temporally local structure in the data while maintaining sparsity of the potential interactions. Our multi-scale framework is based on recursive dyadic partitioning, which recursively partitions the temporal axis into finer intervals and allows us to detect local network structural changes at varying temporal resolutions. The dynamic neighborhood selection is achieved through penalized likelihood estimation, where the penalty seeks to limit the number of neighbors used to model the data. We present theoretical and numerical results describing the performance of our method, which is motivated and illustrated using task-based magnetoencephalography (MEG) data in neuroscience.

Recommendations

• Time-varying estimation and dynamic model selection with an application of network data
• Dynamic chain graph models for time series network data
• Mixed-effect time-varying network model and application in brain connectivity analysis
• Two-way sparsity for time-varying networks with applications in genomics
• Learning non-stationary dynamic Bayesian networks

Cites work

• scientific article; zbMATH DE number 5957408 (Why is no real title available?)
• A theory for multiresolution signal decomposition: the wavelet representation
• Break detection for a class of nonlinear time series models
• CART and best-ortho-basis: a connection
• Causal Network Inference Via Group Sparse Regularization
• Consistency of the group Lasso and multiple kernel learning
• High-dimensional graphs and variable selection with the Lasso
• Investigating Causal Relations by Econometric Models and Cross-spectral Methods
• Model Selection and Estimation in Regression with Grouped Variables
• Multiscale Poisson Intensity and Density Estimation
• Multiscale generalised linear models for nonparametric function estimation
• Multiscale stochastic volatility for equity, interest rate, and credit derivatives.
• Network Granger causality with inherent grouping structure
• Optimal detection of changepoints with a linear computational cost
• Statistical analysis of network data. Methods and models
• Stochastic ordering of multivariate normal distributions
• Unconditional bases are optimal bases for data compression and for statistical estimation

Cited in

(5)

• Time-varying estimation and dynamic model selection with an application of network data
• On the exploration of time-varying networks
• Dynamic chain graph models for time series network data
• Time scale modeling of sparse dynamic networks
• On packing arborescences in temporal networks

This page was built for publication: Dynamic networks with multi-scale temporal structure

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2121707)