Analytics: fairMid
Independent fair-value prices,
built for TCA.
Independent fair-value price series across bonds and IRS, refined continuously as the market moves. Suitable for TCA, valuation, and risk.
What is fairMid
A fair-value reference, drawn from the whole market.
fairMid is an independent fair-value price series for bonds and IRS. It produces a continuous reference series that reflects the market as a whole, rather than any single source.
Its defining property is that the history never stops updating. As more post-trade data lands, sometimes days or weeks after the event, historical fairMid values are refined. The series keeps improving as the picture sharpens.
REGULATOR VIEW
After nearly two years of operating under MiFID II, we are still lacking a reliable view of liquidity across the EU.
– Steven Maijoor, ESMA Chair · MiFID II/MiFIR Review Report (December 2019)
The independent
reference series.
Every feature designed around the needs of TCA, valuation, and risk, three use cases that each require an unconflicted price.
Cross-asset coverage
US Treasuries, UK gilts, European sovereigns, investment-grade and high-yield corporate credit, and cleared IRS. One methodology, one consistent price surface across rates and credit.
BONDS · IRS · CREDIT
A history that refines
fairMid's history never stops updating. As post-trade data continues to land, historical values are refined, so a price for any past date and time improves as the market picture fills in.
CONTINUOUS · POST-TRADE
TCA-ready reference
fairMid is the independent reference against which every fill on Ediphy TCA is measured.
TIMESTAMP · REFERENCE
Valuation-grade output
Consumed as end-of-day marks, period marks, or continuous series. Suitable for FRTB and internal valuation-committee work.
FRTB · VALUATION
The output
A reference price you can consume directly.
Each fairMid response carries the price, the instrument, the timestamp, and a reference to the version of the series under which it was published. Because the history refines as post-trade data lands, a request for a past date and time returns the most refined value available.
Pricing engines, risk systems, and valuation pipelines consume the same series. The output is delivered as an end-of-day mark, a period mark, or a continuous reference, as the consumer requires.
Why does a historical fairMid value change over time? +
Because post-trade data continues to arrive after the event, sometimes days or weeks later. As that information lands, the fair value for a past moment is refined. Each published value carries the version of the series it belongs to, so any change is explicit and traceable, never silent.
Is fairMid tied to any single venue? +
No. fairMid is independent of where any individual trade occurred, reflecting the market as a whole rather than any single source.
How does fairMid differ from liveMid? +
fairMid is a post-trade fair-value series that refines over time as more data lands; liveMid is a real-time mid for trading and live pricing. They serve different purposes and are consumed differently.
What is fairMid suitable for? +
Independent valuation, TCA benchmarking, and risk. It is consumed as end-of-day marks, period marks, or a continuous reference, and supports FRTB and fair-value valuation work.
For quants
An evolving reference, on the record.
An independent fair-value reference is only useful if it reflects the market as completely as possible. Because fairMid refines as post-trade data lands, the value attached to any past moment improves over time, and every value is tied to the version of the series under which it was published.
The questions on the left are the ones we get most often from quant teams during onboarding.
Why Ediphy
A reference drawn
from the whole market.
A price taken from a single source reflects only what that source can see. An independent fair-value reference reflects the market as a whole. fairMid is built on one of the deepest fixed-income datasets in Europe, and it keeps refining: as more post-trade data lands, historical values improve. The breadth of the data and the way it keeps filling in are what make the output trustworthy.