These conditional probabilities demonstrate that each of the partition's four subclasses is causally relevant to patient data <math>D=\{\delta_1,.....\delta_n\}</math> in the population sample <math>PO</math>
These conditional probabilities demonstrate that each of the partition's four subclasses is causally relevant to patient data D = { δ 1 , . . . . . δ n } {\displaystyle D=\{\delta _{1},.....\delta _{n}\}} in the population sample P O {\displaystyle PO}