Full q-scale Esscher positivity and monotonicity from the corrected exit identity
ProvedAvramDividend.Classical.scaleFunction_tilted_positive_monotone_of_corrected_exitavram-classicalcstaresscherfirst-passagemonotonicityscale-functions
Under the standing assumptions and positive discount q, assume the corrected killed two-sided exit formula on 0≤x≤a. Then the q-scale W is strictly positive on the positive axis and admits a positive Esscher tilt parameter for which exp(-φx)W(x) is monotone. The bounded-variation part is canonically Proved unconditionally; the unbounded part follows from the published corrected-exit tilted positivity theorem, with no C1 hypothesis. This result is exactly the accepted original tilted-scale prerequisite with a single additional stochastic exit assumption, suitable for direct import once the corrected exit identity is Proved.
Preamble
import Definitions.Def_AvramDividend_Classical_SpectrallyNegativeLevy import Definitions.Def_AvramDividend_Classical_ScaleFunction import Definitions.Def_AvramDividend_Classical_TwoSidedExitCorrected import Theorems.Thm_AvramDividend_Classical_scaleFunction_tilted_positive_monotone_of_bv import Theorems.Thm_AvramDividend_Classical_unbounded_positive_tilt_of_corrected_exit import Mathlib.Analysis.Complex.Exponential open AvramDividend.Classical MeasureTheory Set open scoped NNReal ENNReal
Formal statement
theorem AvramDividend.Classical.scaleFunction_tilted_positive_monotone_of_corrected_exit
{Ω : Type*} [mΩ : MeasurableSpace Ω]
{P : Measure Ω} {𝓕 : Filtration ℝ≥0 mΩ}
(X : SpectrallyNegativeLevy P 𝓕) (hX : X.Standing)
(q : ℝ) (hq : 0 < q)
(W : ℝ → ℝ) (hW : IsScaleFunction X q W)
(hExit : ∀ a x : ℝ, 0 < a → 0 ≤ x → x ≤ a →
twoSidedExitExpectationCorrected X q a x =
ENNReal.ofReal (W x / W a)) :
(∀ x : ℝ, 0 < x → 0 < W x) ∧
∃ φ : ℝ, 0 < φ ∧
MonotoneOn (fun x : ℝ => Real.exp (-φ * x) * W x) (Ioi (0 : ℝ)) := by sorrySource
Avram Palmowski Pistorius (2007) corrected two-sided exit formula and Esscher scale normalisation; Proved BV renewal and unbounded-exit branches.