The hybrid-to-iterated simplex cell difference is the integrated face defect
ProvedErdos390.WholePaper.BankPaperRealization.fourFiveMovingSimplexHybridCellProductTwo_sub_iteratedanalytic-number-theoryerdos-390erdos390-source-construction
Let be natural numbers with and . Put , , and equal to its real two-variable moving-simplex kernel. The hybrid expression samples the second variable at with weight , whereas the iterated expression integrates both cell variables. Their exact difference is
where is the sum over of the signed Lebesgue cell defects for the conditional moving face at coordinate . Equivalently, its -summand is the right-endpoint approximation minus . This exposes the second-coordinate discretization error without discarding its sign.
Preamble
import Mathlib import Definitions.Def_erdos390_source_bank_interface import Definitions.Def_erdos390_dickman_step_function import Theorems.Thm_MediumPNT import Definitions.Def_erdos390_analytic_foundations_001 import Theorems.Thm_Erdos390_Full_DickmanBasic_kernel_product_bound import Theorems.Thm_Erdos390_Full_DickmanBasic_rho_pos_on_zero_five import Theorems.Thm_Erdos390_Full_DickmanBasic_kernel_secondDerivative_first_bound import Definitions.Def_erdos390_analytic_foundations_009
Formal statement
theorem Erdos390.WholePaper.BankPaperRealization.fourFiveMovingSimplexHybridCellProductTwo_sub_iterated :
∀ {A Y y : Nat} {u : Real},
LE.le.{0} (2 : Nat) y →
LE.le.{0} y A →
Eq.{1}
(HSub.hSub.{0, 0, 0}
(Erdos390.WholePaper.BankPaperRealization.fourFiveMovingSimplexHybridCellProductTwo A Y y u)
(Erdos390.WholePaper.BankPaperRealization.fourFiveMovingSimplexIteratedRealCellProductTwo A Y y u))
(∑ p ∈ Finset.Ioc.{0} A Y,
∫ (x : Real) in ↑(HSub.hSub.{0, 0, 0} p (1 : Nat))..↑p,
HMul.hMul.{0, 0, 0} (Erdos390.WholePaper.BankPaperRealization.fourFiveLogLogLebesgueDensity x)
(Erdos390.WholePaper.BankPaperRealization.fourFiveMovingFaceLebesgueCellDefectSum A Y y u
(Erdos390.WholePaper.BankPaperRealization.fourFiveRealLogCoordinate y x))) := by sorry
Source