Beware of the DAG!

A. Philip Dawid
Proceedings of Workshop on Causality: Objectives and Assessment at NIPS 2008, PMLR 6:59-86, 2010.

Abstract

Directed acyclic graph (DAG) models are popular tools for describing causal relationships and for guiding attempts to learn them from data. They appear to supply a means of extracting causal conclusions from probabilistic conditional independence properties inferred from purely observational data. I take a critical look at this enterprise, and suggest that it is in need of more, and more explicit, methodological and philosophical justification than it typically receives. In particular, I argue for the value of a clean separation between formal causal language and intuitive causal assumptions.

Cite this Paper


BibTeX
@InProceedings{pmlr-v6-dawid10a, title = {Beware of the DAG!}, author = {Dawid, A. Philip}, booktitle = {Proceedings of Workshop on Causality: Objectives and Assessment at NIPS 2008}, pages = {59--86}, year = {2010}, editor = {Guyon, Isabelle and Janzing, Dominik and Schölkopf, Bernhard}, volume = {6}, series = {Proceedings of Machine Learning Research}, address = {Whistler, Canada}, month = {12 Dec}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v6/dawid10a/dawid10a.pdf}, url = {https://proceedings.mlr.press/v6/dawid10a.html}, abstract = {Directed acyclic graph (DAG) models are popular tools for describing causal relationships and for guiding attempts to learn them from data. They appear to supply a means of extracting causal conclusions from probabilistic conditional independence properties inferred from purely observational data. I take a critical look at this enterprise, and suggest that it is in need of more, and more explicit, methodological and philosophical justification than it typically receives. In particular, I argue for the value of a clean separation between formal causal language and intuitive causal assumptions.} }
Endnote
%0 Conference Paper %T Beware of the DAG! %A A. Philip Dawid %B Proceedings of Workshop on Causality: Objectives and Assessment at NIPS 2008 %C Proceedings of Machine Learning Research %D 2010 %E Isabelle Guyon %E Dominik Janzing %E Bernhard Schölkopf %F pmlr-v6-dawid10a %I PMLR %P 59--86 %U https://proceedings.mlr.press/v6/dawid10a.html %V 6 %X Directed acyclic graph (DAG) models are popular tools for describing causal relationships and for guiding attempts to learn them from data. They appear to supply a means of extracting causal conclusions from probabilistic conditional independence properties inferred from purely observational data. I take a critical look at this enterprise, and suggest that it is in need of more, and more explicit, methodological and philosophical justification than it typically receives. In particular, I argue for the value of a clean separation between formal causal language and intuitive causal assumptions.
RIS
TY - CPAPER TI - Beware of the DAG! AU - A. Philip Dawid BT - Proceedings of Workshop on Causality: Objectives and Assessment at NIPS 2008 DA - 2010/02/18 ED - Isabelle Guyon ED - Dominik Janzing ED - Bernhard Schölkopf ID - pmlr-v6-dawid10a PB - PMLR DP - Proceedings of Machine Learning Research VL - 6 SP - 59 EP - 86 L1 - http://proceedings.mlr.press/v6/dawid10a/dawid10a.pdf UR - https://proceedings.mlr.press/v6/dawid10a.html AB - Directed acyclic graph (DAG) models are popular tools for describing causal relationships and for guiding attempts to learn them from data. They appear to supply a means of extracting causal conclusions from probabilistic conditional independence properties inferred from purely observational data. I take a critical look at this enterprise, and suggest that it is in need of more, and more explicit, methodological and philosophical justification than it typically receives. In particular, I argue for the value of a clean separation between formal causal language and intuitive causal assumptions. ER -
APA
Dawid, A.P.. (2010). Beware of the DAG!. Proceedings of Workshop on Causality: Objectives and Assessment at NIPS 2008, in Proceedings of Machine Learning Research 6:59-86 Available from https://proceedings.mlr.press/v6/dawid10a.html.

Related Material