Identifiability by common backdoor in summary causal graphs of time series

TL;DR

This paper investigates the conditions under which causal effects in time series can be identified using common backdoor sets from summary causal graphs, providing algorithms to determine identifiability.

Contribution

It establishes conditions for identifiability by common backdoor sets in time series and offers algorithms to assess this from summary causal graphs.

Findings

Conditions for backdoor identifiability in time series

Algorithms to decide identifiability from summary graphs

Applicability to time series with and without temporal consistency

Abstract

The identifiability problem for interventions aims at assessing whether the total effect of some given interventions can be written with a do-free formula, and thus be computed from observational data only. We study this problem, considering multiple interventions and multiple effects, in the context of time series when only abstractions of the true causal graph in the form of summary causal graphs are available. We focus in this study on identifiability by a common backdoor set, and establish, for time series with and without consistency throughout time, conditions under which such a set exists. We also provide algorithms of limited complexity to decide whether the problem is identifiable or not.