Characterization of causal ancestral graphs for time series with latent confounders

TL;DR

This paper introduces a new class of graphical models for multivariate time series with unobserved confounders, enabling stronger causal inferences without extra assumptions.

Contribution

It characterizes a novel class of causal graphs for time series, showing they are subsets of existing models and provide more causal insight.

Findings

New class of graphs for time series causality

Stronger causal inferences without additional assumptions

Graphical representation of Markov equivalence classes

Abstract

In this paper, we introduce a novel class of graphical models for representing time lag specific causal relationships and independencies of multivariate time series with unobserved confounders. We completely characterize these graphs and show that they constitute proper subsets of the currently employed model classes. As we show, from the novel graphs one can thus draw stronger causal inferences -- without additional assumptions. We further introduce a graphical representation of Markov equivalence classes of the novel graphs. This graphical representation contains more causal knowledge than what current state-of-the-art causal discovery algorithms learn.