Causal inference for transport research
Daniel J. Graham
Abstract
This paper provides a consolidated overview of the statistical literature on causal inference, emphasising its relevance and applicability for transportation research. It outlines a framework for causal identification based on the concept of potential outcomes and provides a summary of core contemporary methods that can be used for estimation. Typical challenges encountered in identifying cause–effect relationships in applied transportation research are analysed via case study simulations, and R code to execute and adapt causal estimators is made available. Causal inference can be used to obtain unbiased and consistent estimates of causal effects in non-experimental settings when interventions or exposures are non-randomly assigned. The paper argues that empirical analyses in transport research are typically conducted in this setting, and consequently, that causal inference has immediate and valuable applicability.