engineering.uncertainty.pooled_sd

O2: s_p = √(Σ ν_i s_i² / Σ ν_i) vs a separate-module identity (≤2 ULP). Not Type A from one sample, not Welch–Satterthwaite ν_eff, and not ISO 21748 u of a mean. ≤2 ULP vs O3 applies only to the published tabulated pooled-standard-deviation vectors (two groups, equal s, sample sizes, an alias, a zero component, and one awkward triple). It is not a whole-domain guarantee and does not cover Type A from one sample, Welch–Satterthwaite ν_eff, or ISO 21748 u of a mean.

This summary records the outcome of the verification run. The full test corpus, oracle implementations, reference tables and generators are retained privately. Their digests identify those artifacts; a digest alone does not let a reader independently reproduce the full verification run.

Recorded result

8/8 passed

Oracle

O3 · mpmath 1.4.1

Reference table commitment (SHA-256)

76a9c81d14e0960762af048916eca26cf23c022e4c4c457166b4d8c615c9f6fd

Generator commitment (SHA-256)

a5b4d7cb19e142ca874153fcb2ab07f6f38ca2c58914b79e83a1927bcf32f7b5

Check a calculation

Use the capability's public input schema and examples to call the versioned REST API or MCP. These interfaces remain available without downloading the validation infrastructure. A successful example checks that example; it does not establish that the complete private suite has been independently reproduced.