lawsynth/docs/methods/causal/sensitivity.md
2026-08-16 22:04:11 +03:00

380 B
Raw Permalink Blame History

Sensitivity summary

e_value accepts a positive finite risk ratio, reciprocates ratios below one, and returns the standard E-value expression RR + sqrt(RR × (RR - 1)), with one mapped to one.

It is a scalar summary of an already supplied risk ratio. It does not estimate the risk ratio, model confounders, or validate the assumptions under which an E-value is meaningful.