Theorem 4.4 — under G ≥ h > 0 and cases (i)–(iii), (4.9) has a unique positive solution and the linear feedback (4.4) is an equilibrium
DisprovedTimeInconsLQ.Deterministic.theorem_4_4Consider the time-inconsistent LQ problem with scalar state () and deterministic coefficients of §4, under the standing assumptions, with control dimension and .
Theorem 4.4. Suppose and one of the following holds on :
- for some , , and for some ;
- for some , , and for some ;
- , for some , and .
Then the coupled Riccati system (4.9) admits a positive solution pair on , unique on . Moreover, let solve the linear ODE (4.8) and set
On every probability space carrying a standard -dimensional Brownian motion and for every initial state, the closed-loop state equation has a solution , and for every such the control is an equilibrium control in the sense of Definition 2.1.
This is the main result of §4: for deterministic coefficients the open-loop equilibrium of the time-inconsistent LQ problem is a linear feedback of the current state, computed from a deterministic system of ODEs.
Formalization Note. Riccati systems are solved in integral form on and uniqueness means agreement on . "Let be a solution of (4.8)" is read as: for every solution. " given by (4.4) is an equilibrium" is read as: a closed-loop state exists, and for every closed-loop state the feedback control satisfies Definition 2.1 (with along sequences, as in the Model module). The paper's Theorem 4.3 proves existence only in case (iii); Theorem 4.4 claims uniqueness in all three cases, and so does this statement.
import Mathlib import Definitions.Def_Peng1990_SMP_Stochastic import Definitions.Def_TimeInconsLQ_Deterministic_Model import Definitions.Def_TimeInconsLQ_Deterministic_Riccati open MeasureTheory ProbabilityTheory Filter Topology Matrix open scoped ENNReal NNReal
namespace TimeInconsLQ.Deterministic
/-- Theorem 4.4 (p. 14). Suppose `G ≥ h > 0` and one of the cases (i), (ii), (iii) holds. Then
(4.9) admits a positive solution pair `(M, N)`, unique on `[0, T]`. Moreover, for every positive
solution pair `(M, N)` and every solution `Φ` of (4.8), on every probability space carrying a
standard `d`-dimensional Brownian motion and for every initial state, the closed-loop equation
of the feedback (4.4) `u*_s = α_sX*_s + β_s` has a solution, and for every such solution `X*`
the control `u*` is an equilibrium (Definition 2.1). -/
theorem theorem_4_4 {Ω : Type*} [MeasurableSpace Ω] {l d : ℕ} (c : Coeffs l d) (T : ℝ≥0)
(hT : 0 < T) (hc : CoeffsStanding c T) (hGh : c.h ≤ c.G) (hh : 0 < c.h)
(hcase : Case_i c T ∨ Case_ii c T ∨ Case_iii c T) :
HasUniqueSolOn T (IsPosSol49 c T) ∧
∀ M N Φ : ℝ≥0 → ℝ, IsPosSol49 c T M N → IsSol48 c T M N Φ →
∀ (P : Measure Ω), IsProbabilityMeasure P →
∀ (W : ℝ≥0 → Ω → Fin d → ℝ) (hW : Peng1990.SMP.IsStdBrownian P W) (x₀ : Fin 1 → ℝ),
(∃ X, IsClosedLoop (toData c P W hW T hT x₀)
(fun s _ => alpha c T M N s) (fun s _ => beta c M Φ s) X) ∧
∀ X, IsClosedLoop (toData c P W hW T hT x₀)
(fun s _ => alpha c T M N s) (fun s _ => beta c M Φ s) X →
TimeInconsLQ.Sufficient.IsEquilibrium (toData c P W hW T hT x₀)
(fun s ω => X s ω 0 • alpha c T M N s + beta c M Φ s) := by sorry
end TimeInconsLQ.Deterministic
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.