Causal Inference by Identification of Vector Autoregressive Processes with Hidden Components

Abstract

A widely applied approach to causal inference from a non-experimental time series $X$, often referred to as "(linear) Granger causal analysis", is to regress present on past and interpret the regression matrix $\hat{B}$ causally. However, if there is an unmeasured time series $Z$ that influences $X$, then this approach can lead to wrong causal conclusions, i.e., distinct from those one would draw if one had additional information such as $Z$. In this paper we take a different approach: We assume that $X$ together with some hidden $Z$ forms a first order vector autoregressive (VAR) process with transition matrix $A$, and argue why it is more valid to interpret $A$ causally instead of $\hat{B}$. Then we examine under which conditions the most important parts of $A$ are identifiable or almost identifiable from only $X$. Essentially, sufficient conditions are (1) non-Gaussian, independent noise or (2) no influence from $X$ to $Z$. We present two estimation algorithms that are tailored towards conditions (1) and (2), respectively, and evaluate them on synthetic and real-world data. We discuss how to check the model using $X$.

References in corpus (2)

• Causal Inference by Identification of Vector Autoregressive Processes with Hidden Components
• Learning the Dependence Graph of Time Series with Latent Factors

Cited by in corpus (7)

• Causal Inference by Identification of Vector Autoregressive Processes with Hidden Components
• Generalization Bounds and Representation Learning for Estimation of Potential Outcomes and Causal Effects
• Tackling the subsampling problem to infer collective properties from limited data
• Fairness in Forecasting and Learning Linear Dynamical Systems
• Recovering the Graph Underlying Networked Dynamical Systems under Partial Observability: A Deep Learning Approach
• Topology Inference for Network Systems with Unknown Inputs
• Topology Inference over Networks with Nonlinear Coupling