Theorem 2, p. 7627 — ρ(E)+ρ(Eᶜ) < 1 violates monotonicity in optimal mixtures on {E,Eᶜ}-measurable acts; the converse holds if max{ρ(E),ρ(Eᶜ)} < 1 or min{ρ(E),ρ(Eᶜ)} = 0
ProvedSurelyBetter.CLB.theorem_2Let be a finite set of states and let be a convex set containing two distinct points; stands for , the range of a nonconstant affine utility on simple lotteries. Acts are represented by their utility profiles (see the definitions item for why this is exact). Let be a c-linearly biseparable preference with capacity : is normalized, monotone and continuous on , and for every event the functional , , represents on all binary acts. Fix an event and write for its complement.
Then:
- If , violates monotonicity in optimal mixtures restricted to acts measurable with respect to .
- If or , and
then satisfies monotonicity in optimal mixtures restricted to acts measurable with respect to .
Here monotonicity in optimal mixtures restricted to a set of acts means: for all with for every state , the set of such that is optimal in is smaller in the set order (Definition 1: iff , imply , ) than the corresponding set for and .
The quantity classifies the attitude toward bets on : below is ambiguity aversion, neutrality, above ambiguity loving. The theorem says that, for the c-linearly biseparable class (which includes maxmin expected utility, Choquet expected utility and -MEU), ambiguity aversion toward even a single event is incompatible with monotonicity in optimal mixtures, while (under the side condition) ambiguity neutrality or loving toward the event is compatible with it.
Formalization Note Acts are utility profiles S → ℝ with values in U; the hypotheses Convex ℝ U and ∃ a ∈ U, ∃ b ∈ U, a ≠ b encode that for a nonconstant affine , and IsCLinBisep U V ρ bundles the standing MBA restriction of §4 with the representation (7) as an order equivalence over binary acts of all events. "The converse" is read, as in the paper's proof, as: implies the property.
import Mathlib import Definitions.Def_Supermodularity_Lattices_InducedSetOrder import Definitions.Def_SurelyBetter_CLB_Model
namespace SurelyBetter.CLB
theorem theorem_2 {S : Type*} [Fintype S] (U : Set ℝ) (hU : Convex ℝ U)
(hU2 : ∃ a ∈ U, ∃ b ∈ U, a ≠ b) (V : (S → ℝ) → ℝ) (ρ : Set S → ℝ)
(hV : IsCLinBisep U V ρ) (E : Set S) :
(ρ E + ρ Eᶜ < 1 → ¬ MOMOn V (measE U E)) ∧
((max (ρ E) (ρ Eᶜ) < 1 ∨ min (ρ E) (ρ Eᶜ) = 0) → 1 ≤ ρ E + ρ Eᶜ →
MOMOn V (measE U E)) := by sorry
end SurelyBetter.CLB
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.