Data-driven covariate selection for nonparametric estimation of causal effects

By Doris Entner, Patrik O. Hoyer, Peter Spirtes. 2013

77 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Causal Inference through a Witness Protection Program (2014, methodology): “Our contribution modifies the results of Entner et al. (2013), who exploit conditional independence constraints to obtain c©2014 Ricardo Silva and Robin Evans. ar X iv :1 40 6.”
A Survey of Learning Causality with Data (2018, background): “s no causal effectt →y. Selection bias can also result from adjustment on certain variables (e.g., colliders or descendants of the outcome variable). Without knowing the complete graph, Entner et al. [41] provide a set of rules that are sufficient to decide whether a set of variables satisfy the bac”
Do-calculus when the True Graph Is Unknown (2015, methodology): “More data-driven methods to find adjustment sets have also been developed (De Luna et al., 2011; Entner et al., 2013; VanderWeele and Shpitser, 2011).”
Matching on Balanced Nonlinear Representations for Treatment Effects Estimation (2017, background): “In recent years, researchers have been exploring the relationships between causal inference and machine learning [39, 10, 38].”
Causal Reasoning and Large Language Models: Opening a New Frontier for Causality (2023, methodology): “In addition to discovering the full graph, techniques that utilize partial graph structure (Entner et al., 2013; Cheng et al., 2022) to identify backdoor sets may also beneﬁt from LLMs’ knowledge-based discovery for inferring speciﬁc edges.”
Ancestral Causal Inference (2016, unknown): “W) =)X6??Y jW [Z Proof. We assume a causal DAG with possible latent variables, the causal Markov assumption, and the causal faithfulness assumption. 1. This is a strengthened version of rule R2(i) in [6]: note that the additional assumptions made there (Y99K6 W, Y99K6 X) are redundant and not actual”