Identification of Non-causal Graphical Models
Preprint
Abstract
The paper considers the problem to estimate non-causal graphical models whose edges encode smoothing relations among the variables. We propose a new covariance extension problem and show that the solution minimizing the transportation distance with respect to white noise process is a double-sided autoregressive non-causal graphical model. Then, we generalize the paradigm to a class of graphical autoregressive moving-average models. Finally, we test the performance of the proposed method through some numerical experiments.Keywords
All Related Versions
This publication has 31 references indexed in Scilit:
- Sparse plus low-rank identification for dynamical latent-variable graphical AR modelsAutomatica, 2023
- Nonparametric identification of Kronecker networksAutomatica, 2022
- Generalized maximum entropy based identification of graphical ARMA modelsAutomatica, 2022
- Learning linear modules in a dynamic network using regularized kernel-based methodsAutomatica, 2021
- Autoregressive identification of Kronecker graphical modelsAutomatica, 2020
- Empirical Bayesian learning in AR graphical modelsAutomatica, 2019
- Identification of Sparse Reciprocal Graphical ModelsIEEE Control Systems Letters, 2018
- Conditional independence graphs for multivariate autoregressive models by convex optimization: Efficient algorithmsSignal Processing, 2017
- Oracle inequalities for high dimensional vector autoregressionsJournal of Econometrics, 2015
- Hellinger Versus Kullback–Leibler Multivariable Spectrum ApproximationIEEE Transactions on Automatic Control, 2008