Identifiability by common backdoor in summary causal graphs of time series
By: Clément Yvernes , Charles K. Assaad , Emilie Devijver and more
Potential Business Impact:
Helps computers learn cause and effect from past data.
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.
Similar Papers
Complete Characterization for Adjustment in Summary Causal Graphs of Time Series
Statistics Theory
Finds hidden causes in past events.
Causal Identification in Time Series Models
Machine Learning (CS)
Finds hidden causes in changing data.
Characterization and Learning of Causal Graphs from Hard Interventions
Machine Learning (Stat)
Finds cause-and-effect relationships from experiments.