English

Causal Vector Autoregression Enhanced with Covariance and Order Selection

Methodology 2022-11-28 v1

Abstract

A causal vector autoregressive (CVAR) model is introduced for weakly stationary multivariate processes, combining a recursive directed graphical model for the contemporaneous components and a vector autoregressive model longitudinally. Block Cholesky decomposition with varying block sizes is used to solve the model equations and estimate the path coefficients along a directed acyclic graph (DAG). If the DAG is decomposable, i.e. the zeros form a reducible zero pattern (RZP) in its adjacency matrix, then covariance selection is applied that assigns zeros to the corresponding path coefficients. Real life applications are also considered, where for the optimal order p1p\ge 1 of the fitted CVAR(p)(p) model, order selection is performed with various information criteria.

Keywords

Cite

@article{arxiv.2211.14203,
  title  = {Causal Vector Autoregression Enhanced with Covariance and Order Selection},
  author = {Marianna Bolla and Dongze Ye and Haoyu Wang and Renyuan Ma and Valentin Frappier and William Thompson and Catherine Donner and Máté Baranyi and Fatma Abdelkhalek},
  journal= {arXiv preprint arXiv:2211.14203},
  year   = {2022}
}
R2 v1 2026-06-28T07:12:54.452Z