representable by a probability content if there is some probability content Q such that every X ∈ X is
integrable w.r.t. Q and Λ(X) = EQ [X].
If X is a Stonean vector lattice, then X ∧ Y = min{X, Y }, X ∨ Y = max{X, Y } ∈ X for X, Y ∈ X in particular
X+ := X ∨ 0, X- := (-X) ∨ 0 ∈ X for any X ∈ X. In this case, if all linear forms from the domain of βρ are
representable by probability contents, then ρ is concentrated on the bounded positions, and as a consequence ρ
admits a robust representation by probability measures if its restriction to the bounded positions does so.
Lemma 6.5 Let X be a Stonean vector lattice, and let every linear form Λ ∈ βρ-1 (R) be representable by some
probability content Q on σ(X). Then we can state:
.1 The sequence ρ(X+ - (X- ∧ n)) n converges to ρ(X) for every X ∈ X.
.2 The sequence ρ((X ∧ m) - Y ) m converges to ρ(X - Y ) for nonnegative X, Y ∈ X, Y being bounded.
.3 inf{∣ρ((X+ ∧ m) — (X- ∧ n)) — ρ(X)∣ ∣ m, n ∈ N} = 0 for X ∈ X, and in addition
sup (-EQ [X] - ρ(X)) = sup (-EQ [X] - ρ(X))
X∈X X∈Xb
for every probability content Q on σ(X) such that each X ∈ X is integrable w.r.t. Q .
.4 If M ⊆ αρ-1 (R) with ρ(X) = sup (—EQ [X] — αρ(Q)) for all bounded X ∈ X, then
Q∈M
ρ(X) = sup (—EQ [X] — αρ (Q)) for all X ∈ X.
Q∈M
Proof:
The most important tool of the proof is Greco’s representation theorem. The reader is kindly referred to [14]
(Theorem 2.10 with Remark 2.3).
Since any Λ ∈ βρ-1 (R) is representable by a probability content, statement .2 follows immediately from Greco’s
representation theorem and Lemma 6.4.
proof of .1:
Let X ∈ X, and let ε > 0. Then there exists some probability content Q with βρ(EQ |X) < ∞ such that the
inequality ρ(X) — ε < —EQ [X+] — βρ(EQ |X) + EQ [X-] holds. Application of Greco’s representation theorem leads
then to
ρ(X) — ε< -Eq[X+] — βρ(EQ∣X)+limEq[X- ∧ n] ≤ limρ(X+ — (X- ∧ n)) ≤ ρ(X)
nn
proof of .3:
Let Q be a probability content on σ(X) such that every X ∈ X is integrable w.r.t. Q, and let X ∈ X. Then for
ε > 0 we may choose by statement .1 and Greco’s representation theorem some n ∈ N with
∣ — Eq [X] — ρ(X) — '—Eq [X+ — (X- ∧ n)] — ρ(X+ — (X- ∧ n))´ ∣ < ɪ
17
More intriguing information
1. Magnetic Resonance Imaging in patients with ICDs and Pacemakers2. Institutions, Social Norms, and Bargaining Power: An Analysis of Individual Leisure Time in Couple Households
3. Estimating the Technology of Cognitive and Noncognitive Skill Formation
4. Portuguese Women in Science and Technology (S&T): Some Gender Features Behind MSc. and PhD. Achievement
5. Heavy Hero or Digital Dummy: multimodal player-avatar relations in FINAL FANTASY 7
6. PER UNIT COSTS TO OWN AND OPERATE FARM MACHINERY
7. Gianluigi Zenti, President, Academia Barilla SpA - The Changing Consumer: Demanding but Predictable
8. Neural Network Modelling of Constrained Spatial Interaction Flows
9. Short Term Memory May Be the Depletion of the Readily Releasable Pool of Presynaptic Neurotransmitter Vesicles
10. The name is absent