docs: fix example variables and intro domain in @stdlib/stats/base/dists/chisquare/{median,mean}

[Planeshifter]

Description

This pull request corrects two documentation-consistency drift findings in @stdlib/stats/base/dists/chisquare — variable-name inconsistencies in median examples and a missing k >= 0 domain qualifier in the mean intro — bringing both packages into line with the established sibling-package conventions in the same namespace. No public behavior, function signatures, or test expectations change.

Namespace summary

• Members analyzed: 14 non-autogenerated packages (cdf, ctor, entropy, kurtosis, logpdf, mean, median, mgf, mode, pdf, quantile, skewness, stdev, variance).
• Features extracted: file tree, package.json shape and contents, manifest.json shape, README headings and prose, test/benchmark/example file naming, public signatures, validation prologues, error-construction style, JSDoc shape and @example variable conventions, TypeScript declaration shape, REPL-help shape, runtime dependencies.
• Clean majority (≥75%): universal package.json metadata (versions, authors, license, engines, OS list, contributors, homepage); universal directory layout per package type; universal JSDoc presence (@param, @returns, @example); var v = <pkg>( ... ); convention across single-argument example blocks (7/8 single-arg siblings); where \k >= 0` is the degrees of freedomintro phrasing within theNonNegativeNumber`-domain group (3/4 packages).
• Features excluded — no clear majority: native binding presence (binding.gyp, lib/native.js, src/, include/, manifest.json, test/test.native.js) splits 10/4 = 71%; factory-function presence (lib/factory.js, test/test.factory.js) splits along semantic lines (functions of an evaluation variable have a factory; pure summary statistics do not); validation predicates split along the genuine mathematical domain of each formula (NonNegativeNumber vs PositiveNumber); test-fixture language differs in stdev (Python) versus the rest (Julia) but fixing it would require regenerating fixtures.

Per outlier package

@stdlib/stats/base/dists/chisquare/median

Fixes variable-name drift in README.md (line 58) and docs/types/index.d.ts (examples 3–4), where the negative-input result was assigned to k instead of v. All seven single-argument sibling packages in this namespace (entropy, kurtosis, mean, mode, skewness, stdev, variance) use var v in the corresponding blocks, as does the package's own lib/main.js JSDoc. Brings conformance to 8/8.

@stdlib/stats/base/dists/chisquare/mean

Corrects the intro formula line from where \k` is the degrees of freedomtowhere `k >= 0` is the degrees of freedom, aligning with the three NonNegativeNumber-domain siblings (mode, stdev, variance) in this namespace and with the Notes block and lib/main.js validation (k < 0returnsNaN`) already present in the same package. The omission of the non-negativity constraint from the intro was an internal consistency drift with no behavioral impact.

Related Issues

None.

Questions

No.

Other

Validation

• Structural feature extraction: filesystem walk over each package's file tree, package.json keys and contents, manifest.json keys, README heading sequences, and test/benchmark/example file naming.
• Semantic feature extraction: one Sonnet agent per package read lib/main.js, lib/index.js, and lib/factory.js (where present) and returned a strict JSON bundle covering public signature, validation prologue, error-construction style, JSDoc shape, TypeScript declaration shape, REPL shape, and runtime dependencies.
• Three-agent drift validation: an Opus semantic-review agent, an Opus cross-reference agent, and a Sonnet structural-review agent each independently inspected the proposed corrections; all three returned confirmed-drift for every item in this PR.
• Deliberately excluded:
  • stdev/test/fixtures/python/* versus the Julia fixtures used by every other package — fixing this would require regenerating fixture data and updating tests, which falls outside the routine's mechanical scope.
  • entropy and kurtosis package.json description fields (Chi-squared distribution entropy. / Chi-squared distribution kurtosis.) versus the broader-stdlib convention of differential entropy / excess kurtosis — ecosystem-wide presence is 62.5% and 85.3% respectively, both below the 90% gate required to flag absence as drift.
  • Native-binding scaffolding (binding.gyp, lib/native.js, src/, include/, manifest.json, benchmark/c/, examples/c/) in ctor, median, pdf, quantile — 71% sibling presence is below the 75% threshold, and adding native bindings would require actual C implementations rather than a mechanical patch. An open PR (feat: add C implementation for stats/base/dists/chisquare/pdf #11634) is separately adding the C implementation for pdf.
  • lib/factory.js, test/test.factory.js — present only in packages that take an evaluation variable (cdf, logpdf, mgf, pdf, quantile), absent in pure summary statistics; this split is semantic, not drift.
  • Validation predicates (k < 0.0 for NonNegativeNumber-domain packages; k <= 0.0 for PositiveNumber-domain packages) — both forms are mathematically correct for the formulas in question.

Checklist

• Read, understood, and followed the contributing guidelines.

AI Assistance

• Yes
• No

If you answered "yes" above, how did you use AI assistance?

• Code generation (e.g., when writing an implementation or fixing a bug)
• Documentation (including examples)
• Research and understanding

Disclosure

This PR was authored by Claude Code as part of a cross-package API drift detection routine: structural and semantic features were extracted from every member of the stats/base/dists/chisquare namespace, candidate drift items were identified by majority vote (≥75% threshold), and each surviving correction was independently confirmed by three validation agents (Opus semantic review, Opus cross-reference, Sonnet structural review) before being applied. The changes are pure documentation-consistency edits — no behavior, signatures, or tests change.

---

---

Generated by Claude Code

[Planeshifter]

/stdlib merge