§5, p. 7625 — given state-by-state monotonicity, independence implies monotonicity in optimal mixtures
ProvedSurelyBetter.CLB.section5_independence_momLet be a finite set of states, a real-valued preference functional on utility profiles , and a convex set of profiles (closed under statewise mixtures). Suppose that on :
- satisfies state-by-state monotonicity (Axiom 2): statewise implies ;
- satisfies independence (Axiom 1): for all and , iff .
Then satisfies monotonicity in optimal mixtures restricted to : for all with statewise,
where is the set of maximizing over and is the set order of Definition 1.
With the set of all profiles this is the paper's statement that, given Axiom 2, independence is a strengthening of monotonicity in optimal mixtures; in particular every subjective expected utility preference satisfies it. In the proof of Theorem 2 it is applied to preferences restricted to acts measurable with respect to .
Formalization Note The paper argues for preferences over all acts; the statement is given for an arbitrary mixture-closed set of profiles, the form used in the converse of Theorem 2. Convexity of keeps the mixtures to which independence is applied inside .
import Mathlib import Definitions.Def_Supermodularity_Lattices_InducedSetOrder import Definitions.Def_SurelyBetter_CLB_Model
namespace SurelyBetter.CLB
theorem section5_independence_mom {S : Type*} [Fintype S] (V : (S → ℝ) → ℝ)
(G : Set (S → ℝ)) (hG : Convex ℝ G) (hmon : StateMonotone V G) (hind : Independence V G) :
MOMOn V G := by sorry
end SurelyBetter.CLB
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.