Yield
Retains whether a rate is executable, indicative, issuer-quoted or realized. A rate kind is part of the number, not footnote decoration.
Fenorium methodology
Comparable does not
mean identical.
Fenorium preserves what makes each instrument different, then gives those differences a consistent place in the comparison.
MARKETS→F→DIMENSIONS
The normalization model
Six fields keep the
comparison honest.
No single field is allowed to stand in for the rest.
%
Duration
Uses time to maturity where it exists. Open-ended instruments require a stated holding period before one-off costs can be annualized.
Liquidity
Separates reported market depth, quoted capacity and executed fills. Missing depth stays missing rather than becoming assumed capacity.
Structure
Records venue, chain, instrument identity, claim type and maturity so a ticker cannot silently collapse distinct contracts.
Quality
Surfaces credit, protocol, peg and oracle observations through a dated rubric. The tier is not a guarantee or external credit rating.
Evidence
Measures whether required inputs are present and supported. It is not a universal risk score, and Fenorium does not currently publish a validated composite Fenorium Score.
From observation to answer
The system can
refuse the question.
Internal decision language lives here—not in the homepage’s first impression.
- 01Resolve identity
Chain, address or venue-specific market identity before comparison.
- 02Classify the rate
Executable, indicative, issuer-quoted, realized or unavailable.
- 03Measure evidence
Maturity, liquidity, underlying quality and declared source gaps.
- 04Apply assumptions
Size, holding period, execution cost and policy haircuts stay explicit.
- 05Return a boundary
A supported answer, a measured rejection or an explicit absence of quote.
BID
The declared inputs support the conditional research decision.
NO BID
A measured issue rejects it under the declared policy.
NO QUOTE
Evidence is insufficient, so no number is manufactured.
Evidence & provenance
A receipt for the method,
not a promise of truth.
Inputs, assumptions and model versions can travel with a decision so it can be inspected or recomputed.
The fetched or quoted field and when it was observed.
What the user or mandate supplied rather than what the engine measured.
Where appropriate, exact bytes can be retained under a content identifier.
One technical mechanism for recording versioned outcomes and visible corrections.
Verification can prove integrity and provenance. It cannot prove that every source observation is objectively true.
Product boundaries
What Fenorium
does not claim.
No guaranteed yield, return or fill.
No claim that reported liquidity equals executable depth.
No custody of user keys or funds.
No assertion that AI creates the underlying number.
No complete live market coverage from a dated snapshot.
No evidence score presented as objective truth.