Identifiability of total effects from abstractions of time series causal graphs

Charles K. Assaad, Emilie Devijver, Eric Gaussier, Gregor Goessler, Anouar Meynaoui
Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, PMLR 244:173-185, 2024.

Abstract

We study the problem of identifiability of the total effect of an intervention from observational time series only given an abstraction of the causal graph of the system. Specifically, we consider two types of abstractions: the extended summary causal graph which conflates all lagged causal relations but distinguishes between lagged and instantaneous relations; and the summary causal graph which does not give any indication about the lag between causal relations. We show that the total effect is always identifiable in extended summary causal graphs and we provide necessary and sufficient graphical conditions for identifiability in summary causal graphs. Furthermore, we provide adjustment sets allowing to estimate the total effect whenever it is identifiable.

Cite this Paper

BibTeX
@InProceedings{pmlr-v244-assaad24a, title = {Identifiability of total effects from abstractions of time series causal graphs}, author = {Assaad, Charles K. and Devijver, Emilie and Gaussier, Eric and Goessler, Gregor and Meynaoui, Anouar}, booktitle = {Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence}, pages = {173--185}, year = {2024}, editor = {Kiyavash, Negar and Mooij, Joris M.}, volume = {244}, series = {Proceedings of Machine Learning Research}, month = {15--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v244/main/assets/assaad24a/assaad24a.pdf}, url = {https://proceedings.mlr.press/v244/assaad24a.html}, abstract = {We study the problem of identifiability of the total effect of an intervention from observational time series only given an abstraction of the causal graph of the system. Specifically, we consider two types of abstractions: the extended summary causal graph which conflates all lagged causal relations but distinguishes between lagged and instantaneous relations; and the summary causal graph which does not give any indication about the lag between causal relations. We show that the total effect is always identifiable in extended summary causal graphs and we provide necessary and sufficient graphical conditions for identifiability in summary causal graphs. Furthermore, we provide adjustment sets allowing to estimate the total effect whenever it is identifiable.} }
Endnote
%0 Conference Paper %T Identifiability of total effects from abstractions of time series causal graphs %A Charles K. Assaad %A Emilie Devijver %A Eric Gaussier %A Gregor Goessler %A Anouar Meynaoui %B Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2024 %E Negar Kiyavash %E Joris M. Mooij %F pmlr-v244-assaad24a %I PMLR %P 173--185 %U https://proceedings.mlr.press/v244/assaad24a.html %V 244 %X We study the problem of identifiability of the total effect of an intervention from observational time series only given an abstraction of the causal graph of the system. Specifically, we consider two types of abstractions: the extended summary causal graph which conflates all lagged causal relations but distinguishes between lagged and instantaneous relations; and the summary causal graph which does not give any indication about the lag between causal relations. We show that the total effect is always identifiable in extended summary causal graphs and we provide necessary and sufficient graphical conditions for identifiability in summary causal graphs. Furthermore, we provide adjustment sets allowing to estimate the total effect whenever it is identifiable.
APA
Assaad, C.K., Devijver, E., Gaussier, E., Goessler, G. & Meynaoui, A.. (2024). Identifiability of total effects from abstractions of time series causal graphs. Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 244:173-185 Available from https://proceedings.mlr.press/v244/assaad24a.html.

Related Material