Tick-Imbalance Bars
Install and import#
npm install fintech-algorithmsimport { constructBars } from "fintech-algorithms/market-data-engineering/bar-construction/tick-imbalance-bars";Signature#
constructBars(trades, config)Closes a bar when signed tick flow becomes unusually one-sided relative to what recent history led you to expect. Bars are emitted on *information* rather than on elapsed time or traded quantity.
Parameters#
| Name | Type | Notes |
|---|---|---|
trades | Trade[] | The raw tape in chronological order. Each trade carries tradeId, timestamp, session, symbol, price, volume and currency. |
config | { initialTickSign: number; initialExpectedTicks: number; initialExpectedTickImbalance: number; alphaTicks: number; alphaTickImbalance: number; thresholdFloor: number; thresholdMultiplier: number; closePartial?: boolean } | The initial* values seed the expectations before any bar has closed; the alpha* values are the EWMA decay rates that update them afterwards. thresholdFloor and thresholdMultiplier bound the resulting threshold so it cannot collapse toward zero in quiet periods. |
Returns#
Bar[] · length fewer
One bar per imbalance event, with the threshold that triggered it recorded on the bar.
Errors#
- When any alpha falls outside 0…1, or a seed expectation is not positive — throws
Complexity: time O(n),
space O(bars).
Worked example#
verified This is the worked example published in the article, replayed by the test suite on every run. The output cannot drift.
Input#
[
{
"tradeId": "W1",
"timestamp": "2026-01-05T14:30:00.000Z",
"session": "2026-01-05",
"symbol": "SYNTH",
"price": 100,
"volume": 10,
"currency": "USD"
},
{
"tradeId": "W2",
"timestamp": "2026-01-05T14:30:01.000Z",
"session": "2026-01-05",
"symbol": "SYNTH",
"price": 100,
"volume": 10,
"currency": "USD"
},
{
"tradeId": "W3",
"timestamp": "2026-01-05T14:30:02.000Z",
"session": "2026-01-05",
"symbol": "SYNTH",
"price": 100.1,
"volume": 10,
"currency": "USD"
}
]Showing 3 of 6 elements.
{
"closePartial": true,
"initialTickSign": 1,
"initialExpectedTicks": 8,
"initialExpectedTickImbalance": 0.5,
"alphaTicks": 0.25,
"alphaTickImbalance": 0.5,
"thresholdFloor": 3,
"thresholdMultiplier": 1
}Call#
constructBars(trades, config)Returns#
array of 1 object
[
{
"barIndex": 0,
"session": "2026-01-05",
"startTime": "2026-01-05T14:30:00.000Z",
"endTime": "2026-01-05T14:30:05.000Z",
"lastTradeTime": "2026-01-05T14:30:05.000Z",
"open": 100,
"high": 100.15,
"low": 100,
"close": 100.15,
"volume": 60,
"dollarValue": 6004,
"tickCount": 6,
"firstTradeId": "W1",
"lastTradeId": "W6"
}
]Diagrams#
Calculation flow#
Causal construction flow
flowchart TD
A["Corrected, eligible, stably ordered trade"] --> B{"New session?"}
B -->|Yes| C["Emit or drop tail; reset price, sign, and expectations"]
B -->|No| D["Keep session state"]
C --> E["Freeze threshold for new bar"]
D --> F["Assign tick sign; flat carries prior sign"]
E --> F
F --> G["Update OHLCV and cumulative tick imbalance"]
G --> H{"Absolute imbalance at least frozen threshold?"}
H -->|No| A
H -->|Yes| I["Emit complete bar and lineage"]
I --> J["Update expected length and expected mean sign"]
J --> E
How it works#
This page states the contract — how to call it correctly. The article explains the concept: why it works, and where it breaks.
References#
- *Advances in Financial Machine Learning*, Section 2.3.2.1 — Marcos López de Prado
- NYSE Daily TAQ Client Specification — New York Stock Exchange, an Intercontinental Exchange company
- Trade Reporting Frequently Asked Questions — Financial Industry Regulatory Authority
- Evidence classification
- Data and licensing note