Market data
Order-book data from decentralized exchanges, and trade and quote history for U.S. stocks and futures, aligned for research and AI.
What is in it
Two bodies of data, held in one research warehouse and read on one clock.
On-chain perpetual futures
Order books rebuilt from snapshots and book diffs, so the full book can be read at any point in its recorded history. Fills, liquidations and funding are stored as their own records alongside the book.
U.S. equities and futures
Trade and quote history from 2018, mapped to one symbol list and aligned in time with the on-chain data.
Record schema
Field names and types for the two core records. Every record carries the source's own timestamp beside our capture time.
Order-book update
| Field | Type | Description |
|---|---|---|
| source_venue | string | Venue that published the update |
| instrument_id | string | Instrument in the Rillor symbol map |
| update_type | enum: snapshot, diff | Full book, or a change to it |
| sequence | integer | Source sequence number, checked for gaps |
| side | enum: bid, ask | Side of the book |
| price | decimal | Price level |
| size | decimal | Size resting at the level after the update; zero removes the level |
| source_ts | UTC timestamp | Event time as the source reports it |
| capture_ts | UTC timestamp | Time the update was captured |
Trade
| Field | Type | Description |
|---|---|---|
| source_venue | string | Venue where the trade printed |
| instrument_id | string | Instrument in the Rillor symbol map |
| trade_id | string | Source trade identifier |
| price | decimal | Trade price |
| size | decimal | Trade size |
| aggressor_side | enum: buy, sell, unknown | Side that initiated the trade, where the source reports it |
| trade_kind | enum: fill, liquidation | Liquidations flagged in on-chain perpetual futures data |
| conditions | list of strings | Sale conditions as reported, for equities and futures |
| source_ts | UTC timestamp | Event time as the source reports it |
| capture_ts | UTC timestamp | Time the trade was captured |
One clock, one symbol map
Data from on-chain venues and from U.S. stocks and futures only becomes useful together when every record sits on the same timeline and every instrument has one name.
- Clock
- Every record is placed on one UTC clock. The source's own timestamp is kept beside the capture time.
- Symbol map
- One identifier per instrument across sources, so a stock, a futures contract and a perpetual on the same underlying can be joined.
- Latency assumptions
- Documented per source: which time each timestamp represents, and the expected delay between the event and its record.
- Sequence checks
- Book sequence numbers are checked on every rebuild, and gaps are written into the coverage notes for that instrument.
What it is used for
Research and AI work that needs real order flow, recorded as it happened.
Forecasting research
Forecasting models tested out of sample on recorded history and scored against baselines.
Forecasting and analyticsCalibrating simulated markets
Spreads, depth and order flow from real books set the conditions AI agents face in simulated markets.
AI evaluation and simulationPrice-discovery studies
How price moves travel between regulated futures and decentralized perpetual futures, measured on one clock.
Backtesting
Recorded books and trades replayed to test models on history they were not built on.
Access
Each part of the data carries the terms of its source.
- DEX order-book data
- Provided under a written data agreement.
- U.S. equities and futures
- Supports research and custom builds within the rights of each source.
- Delivery
- An authenticated API, a scheduled export or an agreed file format.
- Custom builds
- Instruments, history windows and derived tables built to your specification. Custom datasets
Notices
Data rights. Source data is provided within the rights of each source.
Investment. Rillor is not a registered investment adviser or commodity trading advisor and does not provide investment or trading advice. Research, forecasts, data and software described on this site are for research and engineering use. Nothing here is an offer or recommendation to buy or sell any security, commodity interest or digital asset. Past or simulated results do not indicate future results.