Methods for causal discovery
Prominent causal discovery methodologies can be categorized into three types: constraint-based, score-based, and hybrid methods. Each has its strengths and weaknesses and each employs distinct mechanisms for identifying causal structures. Various algorithms have been developed for causal discovery, including the PC algorithm, GES, and the IC algorithm. These algorithms utilize different principles, such as independence tests and scoring functions, to infer causal relationships. The choice of algorithm depends on the nature of the data and the specific requirements of the analysis as each approach has its strengths and weaknesses in uncovering causal structures from observational data. We’ll discuss the methodologies in detail in the following sections.
Constraint-based methods
Researchers Clark Glymour, Peter Spirtes, and Richard Scheines developed constraint-based methods such as the PC algorithm, which uses conditional independence tests...
