Applications of and Tools for Causal Inference Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 19, 2019
  • Saul, Bradley
    • Affiliation: Gillings School of Global Public Health, Department of Biostatistics
  • Various topics related to causal inference with application to infectious disease and ecology are studied and software tools for such applications developed. The causal g-methods of Robins and colleagues – the parametric g-formula, marginal structural models, and structural nested models – are applied to a causal assessment of impaired water quality in North Carolina’s Cape Fear River. The application demonstrates how a potential outcomes’ causal analysis can be done with routine stream monitoring data. Under certain conditions, each of the g-methods can be cast in an estimating equation framework. Causal models often ‘stack’ estimating equations from multiple models, which can be a source of programming errors and bottlenecks. An R package for obtaining point and variance estimates from any arbitrary set of estimating equations is presented. The context of infectious diseases stimulated many advances in causal inference methods in the past 15 years. These methods and important contributions to the science of infectious diseases are reviewed.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Preisser, John
  • Cole, Stephen
  • Stewart, Paul
  • Hudgens, Michael
  • Truong, Kinh
  • Edwards, Jessie
  • Doctor of Public Health
Degree granting institution
  • University of North Carolina at Chapel Hill Graduate School
Graduation year
  • 2017

This work has no parents.