|
Activity Number:
|
563
- Mechanisms of Interference: New Strategies for Identification and Estimation
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 1, 2018 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Social Statistics Section
|
|
Abstract #326536
|
|
|
Title:
|
Estimation of Contagion Effects in Households and Other Networks
|
|
Author(s):
|
Forrest W Crawford* and Wen Wei Loh
|
|
Companies:
|
Yale School of Public Health and Ghent University
|
|
Keywords:
|
interference;
contaigion;
vaccine;
spillover;
causal inference
|
|
Abstract:
|
Estimating the causal effect of an intervention (e.g. vaccination) on infectious disease outcomes is difficult because outcomes may be contagious, and interventions may affect both susceptibility to infection, and infectiousness once infected. We first explain why contagion is fundamentally different from notions of treatment spillover encountered in other work on interference. We then present a significant generalization of a canonical model of contagion to define causal intervention effects for infectious disease interventions in networks. When infection times are observed, we show that these effects are nonparametrically identified under less restrictive assumptions than those typically required by mediation approaches. The causal structure of contagion implies conditional independence relations that simplify identification of causal effects in empirical scenarios. We outline new causal estimands for intervention effects under contagion, and show formally why randomization is not sufficient to eliminate confounding under contagion.
|
Authors who are presenting talks have a * after their name.
