§3.3, p. 6 — the solution of (13) is
ProvedDuffieHedge.Optimal.ode_13_solutionodep2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1real-analysis
Let , let be measurable functions, bounded on , with on . Let be a real function on such that is integrable on and
the integral form of (13). Then
This is the uniqueness of solutions of the linear ODE (13); together with it forces .
Formalization Note. Stated for an arbitrary real function satisfying (13) in integral form, independently of the stochastic model.
Preamble
import Mathlib import Definitions.Def_DuffieHedge_Optimal_Basic open MeasureTheory ProbabilityTheory open scoped ENNReal NNReal
Formal statement
namespace DuffieHedge.Optimal
/-- §3.3, after Lemma 2 (p. 6): the solution of (13). If `m, v` are measurable and bounded on
`[0, T]`, `v` is bounded away from zero, and a real function `H` satisfies (13) in integral form,
`H_t = H_0 − ∫₀ᵗ (m_s²/v_s²) H_s ds` on `[0, T]`, then
`H_t = H_0 exp(−∫₀ᵗ m_s²/v_s² ds)` on `[0, T]`. -/
theorem ode_13_solution (T : ℝ≥0) (m v : ℝ≥0 → ℝ) (hm : Measurable m) (hv : Measurable v)
(hbd : ∃ K : ℝ, ∀ t ≤ T, |m t| ≤ K ∧ |v t| ≤ K)
(hv0 : ∃ δ : ℝ, 0 < δ ∧ ∀ t ≤ T, δ ≤ |v t|)
(H : ℝ≥0 → ℝ)
(hint : IntegrableOn (fun s : ℝ => m s.toNNReal ^ 2 / v s.toNNReal ^ 2 * H s.toNNReal)
(Set.Icc (0 : ℝ) T))
(hode : ∀ t ≤ T, H t = H 0 - ∫ s in Set.Icc (0 : ℝ) t,
m s.toNNReal ^ 2 / v s.toNNReal ^ 2 * H s.toNNReal) :
∀ t ≤ T, H t = H 0 * Real.exp (-∫ s in Set.Icc (0 : ℝ) t,
m s.toNNReal ^ 2 / v s.toNNReal ^ 2) := by sorry
end DuffieHedge.Optimal
Source
Duffie and Richardson, Mean-Variance Hedging in Continuous Time, Ann. Appl. Probab. 1(1) (1991), §3.3, after Lemma 2, p. 6
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.