score +0.2377 sits in the NEUTRAL band [+0.20, +0.40) of the published threshold table ·
snapshot 2026-11-19 17:00 UTC · point-in-time, never edited
Research and education only — not investment advice.
Signal labels are research classifications, not buy/sell recommendations.
Investment implication: none established — no buy, sell, or alpha conclusion is supported by this page.
volume feed not wired — zero-confidence by design (C2 challenger accrues in a parallel shadow table)
C3 Sector velocity
w=0.12
+0.269
C4 Macro regime
w=0.15
+0.600
macro-regime input frozen as of 2026-05-18 with staleness disclosure, pending macro engine restoration
C5 Oil/rates/FX
w=0.05
+0.000
C6 Event impact
w=0.18
-0.446
C7 Peer correlation
w=0.10
+0.303
C8 Cascade effect
w=0.12
+0.000
C9 Model trust
w=0.08
+0.130
Weights are the published composite weights; component scores are
the current snapshot's stored values. C6 event impact carries the highest single weight by design —
evidence is meant to correct price-only signals, not echo them.
Confidence is a model input, not a calibrated probability. Current configuration weights do not establish historical weights. C4 is disclosed as frozen; no past score is changed.
Excerpt corrections and quality flags
Original excerpts and their hashes are never edited; a corrected display excerpt is a separate annotation with its own digest and date. Detector hits stay SUSPECTED until verified against the primary document.
FLAGGED (CONFIRMED) excerpt 3171359e4b40… — verified against the primary document: the excerpt is XBRL context/taxonomy data at its source location, not event-bearing prose; no corrected text exists for this object (rules xbrl_context); original shown unchanged, not yet a correction
Research interpretation and limits
In the 90 days ending 2026-11-19T17:02:03.347145-07:00: 10 event objects from 10 distinct accessions; 0 objects lack an accession. These are descriptive counts, not independent observations.
Composite reconstruction: RECONCILED. Confidence inputs are not calibrated probabilities. Current weights alone cannot establish historical contributions. C4 remains disclosed as frozen from 2026-05-18.
source_hash identifies the original stored event record; it is not a digest of the displayed excerpt or the original SEC file. Public event inclusion verification: NOT_AVAILABLE. /evidence/verify.json describes SIGNAL snapshot ledger checks and does not authenticate this event hash. Quote occurrence, semantic support, arithmetic and external provenance are separate checks.
Evidence coverage — the four terms
95
ECS (0–100)
10
accepted events, 90d
6.1
days since latest event
4
distinct event types
1
substrate active (0/1)
Evidence Coverage v1 (registered protocol): how much evidence stands
under this classification — coverage, not prediction.
Latest accepted events
INSIDER_SELLmagnitude 0.17 · direction -1 ·
4-parsed · 6.1 days ago
George Sarah Catherine S 756 shares @ $1096.01 on 2026-11-12 ($829,065)
EXEC_CHANGEmagnitude 0.80 · direction -1 ·
8-K · 9.1 days ago
On November 7, 2026, Helena B. Foulkes notified the Board of Directors (the “Board”) of Costco Wholesale Corporation (the “Company”) of her decision to resign from the Board and all committees of the Board of which she is a member, effective November 7,
EXEC_CHANGEmagnitude 0.60 · direction +1 ·
8-K · 34.1 days ago
On October 13, 2026, the Compensation Committee of the Board of Directors approved the fiscal 2027 criteria for bonuses to executive officers, with the Nominating and Governance Committee of the Board also approving the environmental and social performance cri
Point-in-time record, oldest → newest; colors follow the label
legend on the home page. Never edited after the fact.
Story membership
No mega-story membership recorded for this name in the current evidence-geometry artifacts.
Research and education only — not investment advice.
Investment implication: none established — no buy, sell, or alpha conclusion is supported by this page. Past results — in-sample or forward-tracked — do not predict future performance.