A two-term distribution bound by the large and small values of h gives an L^p bound for 1 < p < r
ProvedDbarRegularity.lintegral_rpow_le_of_distribution_le_twoharmonic-analysisinterpolationlp-spacesmeasure-theory
Let be an s-finite measure space, let be -a.e. measurable, let be real, let , and suppose that for every
Then
This converts a two-term distribution-function estimate, controlled by the large values of in the first term and by the small values of in the second, into a strong bound with an explicit constant, for every exponent strictly between and .
Formalization Note. are ℝ≥0∞-valued and AEMeasurable; ranges over ℝ≥0 with ; are real; powers are ℝ≥0∞ real powers and integrals are lower Lebesgue integrals ∫⁻. The integrals over and are written with Set.indicator. No finiteness of or of the integrals is assumed; when or is , or , the conclusion may be trivial.
Preamble
import Mathlib.Analysis.SpecialFunctions.Pow.NNReal import Mathlib.MeasureTheory.Integral.Lebesgue.Basic
Formal statement
namespace DbarRegularity
open MeasureTheory
open scoped ENNReal NNReal
/-- Let `g, h ≥ 0` on an s-finite measure space and `1 < p < r`. If for every `t > 0`
`μ{g > t} ≤ (C/t) ∫_{h > t} h + (D/t)^r ∫_{h ≤ t} h^r`, then
`∫ g^p ≤ (C p/(p-1) + D^r p/(r-p)) ∫ h^p`. -/
theorem lintegral_rpow_le_of_distribution_le_two {X : Type*} [MeasurableSpace X]
(μ : Measure X) [SFinite μ] {g h : X → ℝ≥0∞} (hg : AEMeasurable g μ) (hh : AEMeasurable h μ)
{p r : ℝ} (hp : 1 < p) (hpr : p < r) (C D : ℝ≥0∞)
(H : ∀ t : ℝ≥0, 0 < t → μ {x | (t : ℝ≥0∞) < g x} ≤
C / t * ∫⁻ x, {x | (t : ℝ≥0∞) < h x}.indicator h x ∂μ +
(D / t) ^ r * ∫⁻ x, {x | h x ≤ t}.indicator (fun x => h x ^ r) x ∂μ) :
∫⁻ x, g x ^ p ∂μ ≤
(C * ENNReal.ofReal (p / (p - 1)) + D ^ r * ENNReal.ofReal (p / (r - p))) *
∫⁻ x, h x ^ p ∂μ := by sorry
end DbarRegularity
Source
E. M. Stein, Singular Integrals and Differentiability Properties of Functions, Princeton Mathematical Series 30, Princeton University Press (1970), Chapter I, § 4.2, Theorem 5, p. 21