Causal Inference Under Interference And Network Uncertainty

Abstract

Classical causal and statistical inference methods typically assume the observed data consists of independent realizations. However, in many applications this assumption is inappropriate due to a network of dependences between units in the data. Methods for estimating causal effects have been developed in the setting where the structure of dependence between units is known exactly, but in practice there is often substantial uncertainty about the precise network structure. This is true, for example, in trial data drawn from vulnerable communities where social ties are difficult to query directly. In this paper we combine techniques from the structure learning and interference literatures in causal inference, proposing a general method for estimating causal effects under data dependence when the structure of this dependence is not known a priori. We demonstrate the utility of our method on synthetic datasets which exhibit network dependence.

16 pages, published in proceedings of 35th Conference on Uncertainty in Artificial Intelligence (UAI 2019)

References in corpus (7)

• Homophily and Contagion Are Generically Confounded in Observational Social Network Studies
• Extended Bayesian Information Criteria for Gaussian Graphical Models
• A convex pseudo-likelihood framework for high dimensional partial correlation estimation with convergence guarantees
• On Learning Discrete Graphical Models Using Greedy Methods
• A Sound and Complete Algorithm for Learning Causal Models from Relational Data
• Consistent inference of a general model using the pseudo-likelihood method
• Model selection and local geometry

Cited by in corpus (3)

• Estimating Total Treatment Effect in Randomized Experiments with Unknown Network Structure
• Spillover Effects in the Presence of Unobserved Networks
• Identification of causal intervention effects under contagion