Skip to content

Erdős Problem #225

Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: proved (not automatically the live result).
Subject areas: analysis.

Read this status correctly

The status above is reproduced from GCL's immutable external-catalogue snapshot. It is not a new claim of current mathematical resolution, a source-semantic review, an independent proof check, or GCL certification. A published proof and a GCL replay are separate events.

The question in plain language

Consider a trigonometric polynomial of the special form

\[ f(\theta)=\sum_{k=0}^{n}c_k e^{ik\theta}, \qquad \max_{0\le\theta\le2\pi}|f(\theta)|=1. \]

Assume all zeros, interpreted through the corresponding algebraic polynomial \(P(z)=\sum c_kz^k\), lie on the unit circle: their arguments \(\theta\) are real. Erdős asked whether the integral of the modulus satisfies

\[ \int_0^{2\pi}|f(\theta)|\,d\theta\le4. \]

The normalization matters. Replacing the maximum by the average, or allowing algebraic zeros off the unit circle, changes the inequality.

A small example

Set \(P(z)=(1+z)/2\). Then \(f(\theta)=e^{i\theta/2}\cos(\theta/2)\), its unique algebraic root is \(z=-1\), and its maximum modulus is one. Direct integration gives

\[ \int_0^{2\pi}|f(\theta)|\,d\theta =\int_0^{2\pi}|\cos(\theta/2)|\,d\theta=4. \]

Therefore the constant four cannot be reduced. The polynomial \(P(z)=(1+z)^2/4\) has the same zero on the circle with multiplicity two, but its integral is \(\pi\), demonstrating that the constant is not attained by every admissible polynomial.

What is known

PROVED. Saff and Sheil-Small, Coefficient and integral mean estimates for algebraic and trigonometric polynomials with restricted zeros, Journal of the London Mathematical Society (1974/75), proved the statement for complex coefficients. Kristiansen's independent 1974 proof addressed real coefficients; a flaw in an original argument was corrected in his 1976 erratum. It is important to cite the erratum rather than treating all versions of the early proof as equally sound.

The geometric root constraint is the substantial hypothesis. Without it, the trivial normalization gives only \(\int |f|\le2\pi\). In particular, the sharp bound four is not merely the triangle inequality applied coefficient by coefficient.

GCL's current state

The pinned and maintained Erdős #225 classify the original inequality PROVED. Original issue: #1494. The historical complex-coefficient solution and the real-coefficient erratum are source-attributed; their proofs and any upstream formal statement have not been independently replayed.

The original issue contains an authenticated RESULT/1 source report. This is unadjudicated source evidence only, not independent mathematical certification or editorial admission.

What remains

A bounded task evaluates normalized \(P(z)=(1+z)^m/2^m\) for \(m=1,\ldots,8\), integrates its modulus numerically against exact beta-function values, and checks every root lies on \(|z|=1\). Acceptance: the case \(m=1\) returns exactly four, \(m=2\) returns \(\pi\), and any statement-capture ambiguity about “real roots” versus unit-circle roots is resolved against the primary theorem.

Sources and claim boundaries

No new integral inequality, independently checked full source proof or protected editorial admission is claimed.

Catalogue and evidence provenance

  • Pinned catalogue source: teorth/erdosproblems at 536408716724.
  • GCL issue: #1494 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
  • GCL triage: OUTSIDE_STRICT_OPEN_TRIAGE.
  • Formal-statement coverage: No statement was confirmed in GCL's pinned formal-provider snapshot. Presence does not mean the statement or proof has been checked.
  • External mathematical status: consult current problem page and its dated bibliography.

Back to catalogue · Compendium introduction