Absence of arbitrage iff existence of nonnegative state prices
ProvedLinearOptimization.no_arbitrage_iff_state_prices(Theorem 4.8, application of Farkas' lemma to asset pricing) A market operates for a single period in which different assets are traded, with possible states of nature at the end of the period. Investing one dollar in asset pays if the state of nature is , so asset is described by a payoff vector and the market by the payoff matrix .
A portfolio is a vector whose components may be negative (short positions); it yields the state-wealth vector , where
Let be the price of asset and , so acquiring costs .
The absence of arbitrage condition states that any portfolio that pays off nonnegative amounts in every state of nature must have nonnegative cost: if , then .
Theorem 4.8: the absence of arbitrage condition holds if and only if there exists a nonnegative vector such that the price of each asset is given by
import Mathlib.Data.Matrix.Mul
import Mathlib.Data.Real.Basic
open Matrix
/-- **Bertsimas & Tsitsiklis, Theorem 4.8 (p. 168).** No-arbitrage iff state prices exist:
every portfolio `x` with `Rx ≥ 0` costs `p'x ≥ 0` iff there is a
nonnegative state-price vector `q` with `pᵢ = Σ_s q_s r_{si}` for every
asset `i`. -/
theorem LinearOptimization.no_arbitrage_iff_state_prices {m n : ℕ}
(R : Matrix (Fin m) (Fin n) ℝ) (p : Fin n → ℝ) :
(∀ x : Fin n → ℝ, 0 ≤ R.mulVec x → 0 ≤ p ⬝ᵥ x) ↔
∃ q : Fin m → ℝ, 0 ≤ q ∧ ∀ i, p i = ∑ s, q s * R s i := by
sorry
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.