Documentation

TauCeti.Analysis.PDE.Regularity.Hessian.Schauder.WholeSpace

The global Schauder estimate on the whole space #

Let 0 < α < 1, and let L u = ∑ᵢⱼ aᵢⱼ ∂ᵢ∂ⱼu be a non-divergence-form operator on ℝ^{n+1} whose coefficients are uniformly elliptic and bounded with constants 0 < λ ≤ Λ on the whole space (TauCeti.PDE.UniformlyEllipticOn univ a λ Λ) and globally α-Hölder with constant L. This file proves the global Schauder estimate on the whole space: there is a constant C, depending only on n, α, λ, Λ and L, such that every C² function u with bounded, globally α-Hölder Hessian satisfies

‖Du‖_∞ + ‖D²u‖_∞ + [D²u]_α ≤ C ([L u]_α + ‖u‖_∞),

where [·]_α is the global α-Hölder constant (TauCeti.PDE.exists_holderWith_iteratedFDeriv_two_and_norm_le_of_sum_smul_fderiv_fderiv), or in terms of the norm of the bounded space C_b^{2,α}(ℝ^{n+1}), ‖u‖_{C^{2,α}} ≤ C ([L u]_α + ‖u‖_∞) (TauCeti.PDE.exists_norm_le_of_sum_smul_fderiv_fderiv). The Hölder constant of the Hessian enters only qualitatively: the estimate does not depend on it. Unlike the interior estimate on compact sets (PDE.exists_holderOnWith_iteratedFDeriv_two_of_sum_smul_fderiv_fderiv_of_holderOnWith), the constant is uniform over the whole space, and only ‖u‖_∞ remains among the lower-order terms.

The argument #

The interior estimate on small balls (TauCeti.PDE.exists_holderOnWith_iteratedFDeriv_two_closedBall_of_sum_smul_fderiv_fderiv) has a constant that depends on the coefficients only through λ, Λ, L and a bound for their entries, which uniform ellipticity bounds by Λ (TauCeti.PDE.UniformlyEllipticOn.abs_apply_le). The global bounds therefore give one radius R, satisfying the scale condition, and one constant C₁ for balls of radius R about every point. Two points at distance less than R / 2 lie in the half ball about one of them, while at larger distances ‖D²u(x) - D²u(y)‖ ≤ 2N ≤ 2N (R / 2)^(-α) ‖x - y‖^α (HolderOnWith.of_dist_lt_of_dist_le), where N bounds |u|, ‖Du‖ and ‖D²u‖. This is the first estimate, [D²u]_α ≤ C (K + N) (TauCeti.PDE.exists_holderWith_iteratedFDeriv_two_of_sum_smul_fderiv_fderiv_of_holderWith), where K is the Hölder constant of L u.

The first and second derivatives are then removed from the right side by the interpolation inequalities ‖D²u‖_∞ ≤ 32 ε⁻² ‖u‖_∞ + 2 ε^α [D²u]_α and ‖Du‖_∞ ≤ 34 ε⁻¹ ‖u‖_∞ + 2 ε^(1+α) [D²u]_α (HolderWith.norm_iteratedFDeriv_add_two_le_of_norm_le, HolderWith.norm_iteratedFDeriv_succ_le_of_norm_le_of_bddAbove), which for ε ≤ 1 give the common bound (34 ε⁻¹ + 32 ε⁻²) ‖u‖_∞ + 2 ε^α [D²u]_α (HolderWith.norm_iteratedFDeriv_succ_le_and_norm_iteratedFDeriv_add_two_le). Taking for [D²u]_α the Hölder norm nnHolderNorm, which is finite, and ε so small that 4 C ε^α ≤ 1 / 2, the resulting multiple of [D²u]_α is absorbed into the left side.

Main declarations #

References #

theorem TauCeti.PDE.exists_holderWith_iteratedFDeriv_two_of_sum_smul_fderiv_fderiv_of_holderWith {n : ℕ} {α : NNReal} (hα : 0 < α) (hα1 : α < 1) (lam Lam : ℝ) (L : NNReal) :
∃ (C : NNReal), ∀ {a : EuclideanSpace ℝ (Fin (n + 1)) → Matrix (Fin (n + 1)) (Fin (n + 1)) ℝ}, UniformlyEllipticOn Set.univ a lam Lam → (∀ (i j : Fin (n + 1)), HolderWith L α fun (x : EuclideanSpace ℝ (Fin (n + 1))) => a x i j) → ∀ {u : EuclideanSpace ℝ (Fin (n + 1)) → ℝ} {N K : NNReal}, ContDiff ℝ 2 u → (∀ (x : EuclideanSpace ℝ (Fin (n + 1))), |u x| ≤ ↑N) → (∀ (x : EuclideanSpace ℝ (Fin (n + 1))), ‖fderiv ℝ u x‖ ≤ ↑N) → (∀ (x : EuclideanSpace ℝ (Fin (n + 1))), ‖iteratedFDeriv ℝ 2 u x‖ ≤ ↑N) → (∀ (S : Set (EuclideanSpace ℝ (Fin (n + 1)))), IsCompact S → ∃ (V : NNReal), HolderOnWith V α (iteratedFDeriv ℝ 2 u) S) → (HolderWith K α fun (x : EuclideanSpace ℝ (Fin (n + 1))) => ∑ i : Fin (n + 1), ∑ j : Fin (n + 1), a x i j • ((fderiv ℝ (fderiv ℝ u) x) (EuclideanSpace.single i 1)) (EuclideanSpace.single j 1)) → HolderWith (C * (K + N)) α (iteratedFDeriv ℝ 2 u)

The global Schauder estimate with lower-order bounds. Let 0 < α < 1, and let λ, Λ and L be given. Then there is a constant C, depending only on n, α, λ, Λ and L, with the following property. Let a be a matrix-valued coefficient on ℝ^{n+1} which is uniformly elliptic and bounded with constants 0 < λ ≤ Λ (UniformlyEllipticOn univ a λ Λ) and globally α-Hölder entrywise with constant L, and put L u = ∑ᵢⱼ aᵢⱼ ∂ᵢ∂ⱼu. Let u be C², with |u|, ‖Du‖ and ‖D²u‖ bounded by N and with D²u α-Hölder on every compact set (with any constants). If L u is globally α-Hölder with constant K, then D²u is globally α-Hölder with constant C (K + N).

theorem TauCeti.PDE.exists_holderWith_iteratedFDeriv_two_and_norm_le_of_sum_smul_fderiv_fderiv {n : ℕ} {α : NNReal} (hα : 0 < α) (hα1 : α < 1) (lam Lam : ℝ) (L : NNReal) :
∃ (C : NNReal), ∀ {a : EuclideanSpace ℝ (Fin (n + 1)) → Matrix (Fin (n + 1)) (Fin (n + 1)) ℝ}, UniformlyEllipticOn Set.univ a lam Lam → (∀ (i j : Fin (n + 1)), HolderWith L α fun (x : EuclideanSpace ℝ (Fin (n + 1))) => a x i j) → ∀ {u : EuclideanSpace ℝ (Fin (n + 1)) → ℝ} {N K : NNReal}, ContDiff ℝ 2 u → (∀ (x : EuclideanSpace ℝ (Fin (n + 1))), |u x| ≤ ↑N) → BddAbove (Set.range fun (x : EuclideanSpace ℝ (Fin (n + 1))) => ‖iteratedFDeriv ℝ 2 u x‖) → MemHolder α (iteratedFDeriv ℝ 2 u) → (HolderWith K α fun (x : EuclideanSpace ℝ (Fin (n + 1))) => ∑ i : Fin (n + 1), ∑ j : Fin (n + 1), a x i j • ((fderiv ℝ (fderiv ℝ u) x) (EuclideanSpace.single i 1)) (EuclideanSpace.single j 1)) → HolderWith (C * (K + N)) α (iteratedFDeriv ℝ 2 u) ∧ (∀ (x : EuclideanSpace ℝ (Fin (n + 1))), ‖fderiv ℝ u x‖ ≤ ↑C * (↑K + ↑N)) ∧ ∀ (x : EuclideanSpace ℝ (Fin (n + 1))), ‖iteratedFDeriv ℝ 2 u x‖ ≤ ↑C * (↑K + ↑N)

The global Schauder estimate on the whole space. Let 0 < α < 1, and let λ, Λ and L be given. Then there is a constant C, depending only on n, α, λ, Λ and L, with the following property. Let a be a matrix-valued coefficient on ℝ^{n+1} which is uniformly elliptic and bounded with constants 0 < λ ≤ Λ (UniformlyEllipticOn univ a λ Λ) and globally α-Hölder entrywise with constant L, and put L u = ∑ᵢⱼ aᵢⱼ ∂ᵢ∂ⱼu. Let u be C² with |u| ≤ N, and with D²u bounded and globally α-Hölder (with any constant). If L u is globally α-Hölder with constant K, then D²u is globally α-Hölder with constant C (K + N), and ‖Du‖ and ‖D²u‖ are bounded by C (K + N).

theorem TauCeti.PDE.exists_norm_le_of_sum_smul_fderiv_fderiv {n : ℕ} {α : NNReal} (hα : 0 < α) (hα1 : α < 1) (lam Lam : ℝ) (L : NNReal) :
∃ (C : NNReal), ∀ {a : EuclideanSpace ℝ (Fin (n + 1)) → Matrix (Fin (n + 1)) (Fin (n + 1)) ℝ}, UniformlyEllipticOn Set.univ a lam Lam → (∀ (i j : Fin (n + 1)), HolderWith L α fun (x : EuclideanSpace ℝ (Fin (n + 1))) => a x i j) → ∀ (u : CkHolderSpace 2 α (EuclideanSpace ℝ (Fin (n + 1))) ℝ) {N K : NNReal}, (∀ (x : EuclideanSpace ℝ (Fin (n + 1))), |u x| ≤ ↑N) → (HolderWith K α fun (x : EuclideanSpace ℝ (Fin (n + 1))) => ∑ i : Fin (n + 1), ∑ j : Fin (n + 1), a x i j • ((fderiv ℝ (fderiv ℝ ⇑u) x) (EuclideanSpace.single i 1)) (EuclideanSpace.single j 1)) → ‖u‖ ≤ ↑C * (↑K + ↑N)

The global Schauder estimate in the C_b^{2,α} norm. Let 0 < α < 1, and let λ, Λ and L be given. Then there is a constant C, depending only on n, α, λ, Λ and L, with the following property. Let a be a matrix-valued coefficient on ℝ^{n+1} which is uniformly elliptic and bounded with constants 0 < λ ≤ Λ (UniformlyEllipticOn univ a λ Λ) and globally α-Hölder entrywise with constant L, and put L u = ∑ᵢⱼ aᵢⱼ ∂ᵢ∂ⱼu. Then every u ∈ C_b^{2,α}(ℝ^{n+1}) with |u| ≤ N, for which L u is globally α-Hölder with constant K, satisfies ‖u‖_{C^{2,α}} ≤ C (K + N).