evpi

voi.evpi

Expected value of perfect information from the probabilistic analysis.

Functions

Name	Description
evpi	Expected value of perfect information per decision.

evpi

voi.evpi.evpi(outcomes, wtp, *, effect=None)

Expected value of perfect information per decision.

EVPI is the expected NMB gain from resolving all uncertainty: E[max_d NMB_d] - max_d E[NMB_d]. It equals the smallest expected loss across interventions at each willingness-to-pay value, so it reads straight off the expected loss curves.

Parameters

Name	Type	Description	Default
outcomes	Outcomes	Outcomes from a probabilistic sensitivity analysis.	required
wtp	float | ArrayLike | Sequence[float]	A willingness-to-pay value or grid.	required
effect	str | None	Effect column (default: the primary effect).	None

Returns

Name	Type	Description
	float | pd.Series	A float for scalar wtp; a Series indexed by wtp for a grid.

Example

import pandas as pd from heormodel.models import Outcomes from heormodel.voi import evpi c = pd.DataFrame({“A”: [0.0, 0.0], “B”: [5.0, 5.0]}) e = pd.DataFrame({“A”: [0.0, 0.0], “B”: [1.0, -1.0]}) evpi(Outcomes.from_wide(c, e), wtp=10.0) 2.5