Inference of High-dimensional Autoregressive Generalized Linear Models

Abstract

Vector autoregressive models characterize a variety of time series in which linear combinations of current and past observations can be used to accurately predict future observations. For instance, each element of an observation vector could correspond to a different node in a network, and the parameters of an autoregressive model would correspond to the impact of the network structure on the time series evolution. Often these models are used successfully in practice to learn the structure of social, epidemiological, financial, or biological neural networks. However, little is known about statistical guarantees on estimates of such models in non-Gaussian settings. This paper addresses the inference of the autoregressive parameters and associated network structure within a generalized linear model framework that includes Poisson and Bernoulli autoregressive processes. At the heart of this analysis is a sparsity-regularized maximum likelihood estimator. While sparsity-regularization is well-studied in the statistics and machine learning communities, those analysis methods cannot be applied to autoregressive generalized linear models because of the correlations and potential heteroscedasticity inherent in the observations. Sample complexity bounds are derived using a combination of martingale concentration inequalities and modern empirical process techniques for dependent random variables. These bounds, which are supported by several simulation studies, characterize the impact of various network parameters on estimator performance.

Submitted to IEEE Transactions on Information Theory

References in corpus (3)

• Forecasting emergency medical service call arrival rates
• Scalable Inference for Neuronal Connectivity from Calcium Imaging
• Bayesian Inference of Whole-Brain Networks

Cited by in corpus (14)

• Connecting the Dots: Identifying Network Structure via Graph Signal Processing
• Neural Granger Causality
• Multi-sample Estimation of Bacterial Composition Matrix in Metagenomics Data
• Learning Markov models via low-rank optimization
• Majorizing Measures, Sequential Complexities, and Online Learning
• A Survey of Estimation Methods for Sparse High-dimensional Time Series Models
• Structural Break Detection in High-Dimensional Non-Stationary VAR models
• Estimating Network Structure from Incomplete Event Data
• Context-dependent self-exciting point processes: models, methods, and risk bounds in high dimensions
• Improved Prediction and Network Estimation Using the Monotone Single Index Multi-variate Autoregressive Model
• Near-optimal Offline and Streaming Algorithms for Learning Non-Linear Dynamical Systems
• High-Dimensional Bernoulli Autoregressive Process with Long-Range Dependence
• Exponential Family Graphical Models: Correlated Replicates and Unmeasured Confounders, with Applications to fMRI Data
• High-Dimensional Dynamic Systems Identification with Additional Constraints