Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Poisson-Jensen inequality for Hardy functions

Statement

Let 0<p≤∞ and f∈Hp(D) with f≢0, with boundary function f∗∈Lp as in Boundary values and log-integrability of Nevanlinna-class functions; by Log-integrability of the boundary values of a Hardy function one has log⁡∣f∗∣∈L1(T,m). Then for every z∈D log⁡∣f(z)∣≤P[log⁡∣f∗∣](z):=∫TP(z,ζ) log⁡∣f∗(ζ)∣ dm(ζ), with the convention log⁡0=−∞. In particular log⁡∣f∣ is majorized on D by the harmonic function P[log⁡∣f∗∣], and equality holds for every z whenever f is outer in the sense of Inner, singular inner and outer functions.

Facts & Assumptions

Given: Countable choice, 0<p≤∞, a nonzero f∈Hp(D), its boundary function f∗ with log⁡∣f∗∣∈L1(T,m), a point z∈D with f(z)≠0, and radii R∈(∣z∣,1).

[L1]

A nonzero Hardy function lies in N and has finite nonzero nontangential boundary values under countable choice, with f∗∈Lp and log⁡∣f∗∣∈L1. In particular radial limits exist almost everywhere. The elementary estimate log⁡+t≤tq/q holds for every q>0. No strong Lp convergence is assumed in this proof. (Boundary values and log-integrability of Nevanlinna-class functions, Log-integrability of the boundary values of a Hardy function, Analytic Hardy spaces on the unit disc, The Nevanlinna class on the disc, The Axiom of Countable Choice (ACω))

[L2]

fR(w):=f(Rw) is holomorphic on a neighbourhood of the closed unit disc, and F(w):=fR(φz(w)) likewise, where φz(w)=z−w1−z‾w is the Blaschke factor; F(0)=f(Rz) and F has finitely many zeros in the open disc (The unit disc, the upper half-plane, and Blaschke factors, Blaschke factors are automorphisms of the disc).

[L3]

Jensen's formula: for F holomorphic on a neighbourhood of the closed unit disc with F(0)≠0, log⁡∣F(0)∣=12π∫02πlog⁡∣F(eiθ)∣ dθ−∑jlog⁡1∣wj∣, the sum over the zeros wj of F in the open unit disc with multiplicity; if F meets a boundary zero the identity is recovered by limits (Jensen's formula on a disc).

[L4]

Change of variables on the circle: φz maps T bijectively onto itself with ∣φz′(ζ)∣=1−∣z∣2∣1−z‾ζ∣2=P(z,ζ), so 12π∫02πG(φz(eiθ)) dθ=∫TG(ζ)P(z,ζ) dm(ζ) for every integrable G on T (Blaschke factors are automorphisms of the disc, The Poisson kernel on the unit disc, The unit disc, the upper half-plane, and Blaschke factors, Conjugation is an involutive real-field automorphism, zz‾=∣z∣2, and modulus is definite, multiplicative, and subadditive).

[L5]

For an L1 real datum its Poisson integral is harmonic; the Poisson kernel has unit mass and is a bounded positive continuous weight at each fixed interior z. Dominated convergence applies to bounded truncated logarithms, and Fatou's lemma to their nonnegative weighted negative parts. (The Poisson integral of a finite complex boundary measure, The Poisson kernel is positive, has total mass one, and concentrates at a boundary point, Dominated convergence, Fatou's lemma)

[L6]

A holomorphic f is outer exactly when log⁡∣f(z)∣=P[log⁡∣f∗∣](z) for all z (Inner, singular inner and outer functions).

Proof

technique · direct
1.1givenL2L3L4algebra

The Jensen inequality. Assume first f(z)≠0 and take R<1 sufficiently close to 1 that f(Rz)≠0, which follows from continuity and f(z)≠0. By [L2] the function F(w)=fR(φz(w)) is holomorphic on a neighbourhood of the closed unit disc with F(0)=f(Rz)≠0; applying [L3] and dropping the nonnegative zero terms gives log⁡∣f(Rz)∣≤12π∫02πlog⁡∣fR(φz(eiθ))∣ dθ (if F has boundary zeros, use the limiting form of [L3]). By the change of variables [L4], the right side equals ∫Tlog⁡∣fR(ζ)∣P(z,ζ) dm(ζ)=∫TP(z,ζ)log⁡∣f(Rζ)∣ dm(ζ).

1.2givenalgebra

The case f(z)=0. If f(z)=0, then log⁡∣f(z)∣=−∞≤P[log⁡∣f∗∣](z) by the convention on log⁡0; the inequality holds trivially.

1.3L1L5givenalgebra

Positive logarithmic parts converge in L1. For finite p set C=∥f∥Hpp, so ∫∣fR∣pdm≤C and ∫∣f∗∣pdm≤C by [L1]. Given ε>0, choose L>0 large enough that log⁡t≤εtp for all t>eL; indeed [L1] with q=p/2 gives log⁡t/tp≤(2/p)t−p/2 for t≥1, and it suffices to take L with (2/p)e−Lp/2≤ε. Writing uR=log⁡+∣fR∣, u∗=log⁡+∣f∗∣, the tails satisfy ∫(uR−L)+dm≤εC,∫(u∗−L)+dm≤εC. The truncated functions min⁡(uR,L) converge almost everywhere to min⁡(u∗,L) and are bounded by L, so [L5] gives L1 convergence by dominated convergence. Therefore lim sup⁡R↑1∥uR−u∗∥1≤2εC, and letting epsilon decrease to zero proves the claim. For p infinity, all positive logarithms are bounded by log⁡+∥f∥∞, so dominated convergence applies directly. These limit statements hold along every sequence tending to one, hence for the stated radial limit.

2.1step 1.1step 1.2step 1.3L1L5algebra

Pass to the Jensen inequality. For f(z) nonzero, let R increase to one in step 1.1. Its left side tends to log⁡∣f(z)∣. The weight P(z,⋅) is bounded by [L5], so step 1.3 gives convergence of the weighted positive logarithmic integrals. Fatou's lemma gives ∫P(z,ζ)log⁡−∣f∗(ζ)∣dm≤lim inf⁡R↑1∫P(z,ζ)log⁡−∣f(Rζ)∣dm. Subtracting this inequality from the positive-part limit bounds the limsup of the Jensen right side by P[log⁡∣f∗∣](z). This quantity is finite by [L1] and the bounded weight, so log⁡∣f(z)∣≤P[log⁡∣f∗∣](z). Step 1.2 covers zeros of f. Only CC boundary existence and the logarithmic tail estimate were used; no AC Hardy representation is invoked.

3.1step 1.1step 2.1L6algebra∎

Equality for outer functions. If f is outer, then by [L6] the identity log⁡∣f(z)∣=P[log⁡∣f∗∣](z) holds for every z, so equality holds in the Poisson-Jensen inequality; thus the inequality is an identity for all nonzero outer functions.

Depends on

Used by

Dependency tree · two levels

78 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources