Causal Graphical Models and Their Applications

The study of causality examines the relations and regularities found in a given environment that are stronger than merely probabilistic relations, in the sense that a causal relation allows for evaluating a change in the consequence given a change in the cause. Causation is crucial not only for predictive accuracy but also for control, explanation, and intervention. Causality is central to much of the scientific enterprise-and even everyday life-and there has been increasing interest in recent decades in causal modeling, inference, and reasoning. Causal models enable us to perform other types of reasoning that are not possible with associative models: (i) interventions, where we want to find the effects of an external agent forcing a variable to a specific value (in contrast with merely observing that variable's value); and (ii) counterfactuals, where we want to reason about what would have happened if certain information had been different from what actually happened. Causality usually cannot be directly observed, and so a long-standing goal of causality researchers has been causal discovery: inferring the underlying causal model from observational data. Multiple causal discovery methods have been developed and successfully applied in different fields, such as biology, medicine, and economics, among others. This Special Issue presents recent advances in causal reasoning and causal discovery based on causal graphical models, including novel applications in different domains. It includes contributions from theory to applications in the following aspects: theory, causal reasoning, causal discovery, and applications.