Guard bands and one-sided acceptance
Research context
research method · conditional · As of 2026-10-10
For an upper specification limit T and assigned Y|data∼N(y,u²), accepting only y≤T−zu ensures Pr(Y>T|data)≤1−Φ(z) under that model.
Key knowledge
For an upper specification limit T and assigned Y|data∼N(y,u²), accepting only y≤T−zu ensures Pr(Y>T|data)≤1−Φ(z) under that model. A positive guard band shrinks acceptance; it trades lower false-acceptance risk against more false rejections. A guarded-rejection rule trades risks differently and must name its rejection and indeterminate regions.
Values and equations
A=T−zu; accept if y≤A; Pr(Y>T|data)≤1−Φ(z)
- A,T,y,u
- Acceptance limit, specification, estimate and standard uncertainty; same units
- z
- One-sided standard-normal quantile
- Φ
- Standard-normal cumulative probability
Conditions and scope
- Normal assigned distribution, known u>0 and upper-limit rule; Φ is the standard normal CDF. Other distributions need their own quantile.
- This is a specific-item posterior/assigned risk statement, not a global consumer-risk guarantee. An arbitrary k=2 band does not automatically encode a chosen loss function.
Sources · 1
References
Sources retain their recorded scope. A source link alone does not establish verification of every statement.
Evidence and review
Inspected §§7–9: conformity probability, guarded acceptance/rejection and specific/global risks; source check dated 2026-10-10. Scope is limited to the cited statements, equations and assumptions, not the whole publication.
2026-10-10 · Author source inspection of the displayed claim, model assumptions and equations; independent exact-content and branch-inventory review is a separate gate. Contemporary status is a dated snapshot, not a promise of ongoing monitoring.
Entry details
Permanent link to this entry ↗- Record identity
- research-measurement-guard-band-decision
- Revision
- research.1
- Structured reference record
- Postgraduate & research reference