LIVE COVERAGEMSCI EM1,084.22+0.42%BRENT82.14−0.18%DXY103.81+0.06%CN 10Y2.241−0.03ENTITIES INDEXED500M+
EN·CLIENT PORTAL
§ 03 · Methodology
StandardTier-1 official first
VerificationHuman-in-loop · 14 analysts
ProvenancePer-series, per-call

Acquisition. Resolution.
Verification. Delivery.

Every series in EDS carries explicit provenance — source tier, transform note, refresh window, and confidence score. Where official data is missing or late, the portal labels the series with a caveat and exposes the fallback logic. We publish what we know and how we know it.

§ 01 · The four-stage pipeline

From raw signal to defensible series.

Data flows through four stages. Each stage emits a versioned artifact with full lineage. If a downstream stage rejects an upstream artifact, we publish the rejection — not a silent retry. Transparency is the architecture, not a feature on top.

01

Acquisition

Tier-1 official sources first — statistics offices, regulators, exchange feeds. Secondary sources and OSINT explicitly tagged with their tier and confidence ceiling.

3,200+ feeds · daily
02

Resolution

Proprietary entity-resolution engine disambiguates duplicates and unifies identifiers across jurisdictions. Confidence scored per identifier, never per record.

Confidence ≥ 0.92
03

Verification

Schema, null, dup, and freshness guards. Structural breaks reviewed by a regional analyst before release. Audit trail attached.

Human-in-loop · 14 analysts
04

Delivery

API, portal, and bulk export — each return source, last-updated, unit, and caveat badges per series. No black-box magic.

SLA: 99.9% uptime
§ 02 · Source hierarchy

What counts as a source, and how we rank it.

Not all sources are equal. We rank every feed on a four-tier scale, and the tier is exposed on every series. Tier-1 sources earn an unconditional confidence ceiling of 1.00; tier-4 sources are capped at 0.70 by policy regardless of volume.

Tier Source class Examples Confidence ceiling Refresh
Tier IOfficial primaryStatistics offices, regulators, exchange feeds, central banks1.00≤ 4h
Tier IICurated secondaryMultilateral institutions, audited industry bodies, vetted commercial0.95≤ 12h
Tier IIIVerified OSINTLeak archives (cross-confirmed), trade data, satellite-derived series0.85≤ 24h
Tier IVUnverified OSINTSingle-source field reports, investigative tips, single-feed signals0.70 (cap)on event
§ 03 · Confidence scoring

Every series carries a confidence score.

Confidence is a function of source tier, recency, cross-confirmation count, and historical accuracy of the series. It is computed at ingestion, recomputed on every transform, and decays with staleness.

w₁Source tier weight (0.40)tier-bound
w₂Cross-confirmation count (0.25)log-scaled
w₃Historical accuracy of series (0.20)backtested
w₄Recency decay (0.15)half-life: tier
Confidence score0.00 – 1.00
EDS / CONFIDENCE · v3.4 SAMPLE SERIES · ENT-44219 RECOMPUTED ON EVERY TRANSFORM ε = 0.04 1.000.85 0.700.50 T-12MT-6MNOW 0.96
§ 04 · Caveats & corrections

What we won't pretend.

We publish a public corrections log and a per-series caveat badge. Coverage is uneven across the world by design — we go deep where institutional decisions are hardest to make, and we say so plainly where we don't.

Caveat: stale

Refresh past window

Series held back when source feed is past its expected refresh by 2× the median interval. Confidence decays automatically until refreshed.

Caveat: thin

Single-source coverage

Series with only one Tier I/II source carry a thin badge and a confidence cap of 0.85 regardless of accuracy.

Caveat: contested

Sources disagree

When two Tier I sources disagree by > 10%, both are surfaced with a contested badge. We do not silently pick a winner.

Documentation

Read the methodology in full.

The full methodology document, including resolver pseudocode and confidence math, is available under NDA. Briefings include a walkthrough with a senior analyst.