Ordinal Causal Discovery. (arXiv:2201.07396v1 [stat.ME])

Causal discovery for purely observational, categorical data is a
long-standing challenging problem. Unlike continuous data, the vast majority of
existing methods for categorical data focus on inferring the Markov equivalence
class only, which leaves the direction of some causal relationships
undetermined. This paper proposes an identifiable ordinal causal discovery
method that exploits the ordinal information contained in many real-world
applications to uniquely identify the causal structure. Simple score-and-search
algorithms are developed for structure learning. The proposed method is
applicable beyond …

arxiv discovery