Positivity of the futures price: a.s. for every
OpenDuffieHedge.Optimal.F_posWork in the futures-hedging market of Duffie and Richardson under the standing hypotheses of §3.1: the futures price solves the linear stochastic differential equation (1)
where , the coefficients are measurable and bounded on , is bounded away from , and . Then the futures price is almost surely strictly positive at every time:
This is the positivity of the geometric (log-normal type) price process (1), . The paper uses it implicitly whenever it divides by , in particular in the feedback rule (11) and in the identity used in the proof of Lemma 2.
Formalization Note. is only assumed to be an Itô process solving (1) in the sense of Peng1990.SMP.IsItoProcess (part of Market.Standing); positivity is not built into the market structure, so it is stated here as a consequence of the standing hypotheses.
import Mathlib import Definitions.Def_DuffieHedge_Optimal_Basic open MeasureTheory ProbabilityTheory open scoped ENNReal NNReal
namespace DuffieHedge.Optimal
/-- (1): under the standing hypotheses the futures price is almost surely strictly positive at
every time `t ∈ [0, T]` (it is the geometric solution of `dF = mF dt + vF dξ` with `F₀ > 0`). -/
theorem F_pos {Ω : Type*} [MeasurableSpace Ω] (M : Market Ω) (hM : M.Standing) :
∀ t ≤ M.T, ∀ᵐ ω ∂M.P, 0 < M.F t ω := by sorry
end DuffieHedge.Optimal