An integrated approach to casuality: the role of casual graphs

Causal questions are central for most biomedical and social science studies. The mainframeworks that allow the analysis of causal relations are Potential Outcomes andCausal Graphs. The approaches have often been compared, contrasting their relativestrengths. This paper evaluates the implications of merging the two methodologies inan integrated approach. In particular, we assess how the limits of one can be compen-sated by the solutions provided by the other. The outlined approach employs causalgraphs to discover and formalize a causal model that is then used as a guide to imple-menting potential outcomes identification strategies. The integrated approach could bebeneficial to both frameworks. The assumptions of potential outcome methods can beassessed directly from a causal graph even in high dimensional contexts, thus makingthe obtained causal estimates more reliable. On the other hand, causal graphs can ben-efit from the several ad hoc identification strategies that have been developed in thepotential outcomes literature.