Neural Additive Vector Autoregression Models for Causal Discovery in Time Series

TL;DR

This paper introduces Neural Additive Vector Autoregression (NAVAR), a neural network-based method for discovering nonlinear causal relationships in multivariate time series data, outperforming existing approaches on benchmark datasets.

Contribution

NAVAR is a novel neural approach that captures nonlinear causal relations in time series, providing interpretable causal influence measures and state-of-the-art performance.

Findings

Achieves state-of-the-art results on benchmark datasets

Effectively discovers nonlinear causal relationships

Provides interpretable causal influence mappings

Abstract

Causal structure discovery in complex dynamical systems is an important challenge for many scientific domains. Although data from (interventional) experiments is usually limited, large amounts of observational time series data sets are usually available. Current methods that learn causal structure from time series often assume linear relationships. Hence, they may fail in realistic settings that contain nonlinear relations between the variables. We propose Neural Additive Vector Autoregression (NAVAR) models, a neural approach to causal structure learning that can discover nonlinear relationships. We train deep neural networks that extract the (additive) Granger causal influences from the time evolution in multi-variate time series. The method achieves state-of-the-art results on various benchmark data sets for causal discovery, while providing clear interpretations of the mapped causal relations.