Data is the foundation

Every backtest and every live signal is only as good as the data underneath it. These pages explain the data an algorithmic trader works with — bars, ticks, corporate-action adjustments — and the unglamorous but decisive craft of cleaning, validating and aligning it, because a single bad price or a missed split can silently wreck a strategy.

Market Data: Market data is the raw material of algorithmic trading: OHLC bars, tick-by-tick trades and quotes, and minute bars, plus reference data like corporate actions. Using it correctly means adjusting prices for splits and dividends, cleaning errors and outliers, handling missing data, aligning time zones and timestamps, and validating everything — because garbage data produces convincing but false backtests and dangerous live signals.

OHLC Data

Bar data

OHLC data summarises a time interval into four prices — the open, the high, the low and the close — giving a compact bar that records where price sta…

→

Tick Data

Tick data

Tick data is the finest-grained market record — every individual trade and often every quote update — each stamped with a price, a size and a precise…

→

Minute Data

Intraday bars

Minute data is intraday OHLC bars sampled at one-minute (or a few-minute) intervals, giving enough intraday resolution for most systematic strategies…

→

Corporate Actions

Reference data

Corporate actions are company events — splits, bonuses, dividends, rights issues and mergers — that mechanically change a stock's price or share coun…

→

Data Cleaning

Data quality

Data cleaning is the systematic detection and correction of errors in market data — bad prints, outliers, duplicates and spurious spikes — so that a …

→

Missing Data

Data quality

Missing data is the absence of expected observations in a time series — gaps from holidays, trading halts, illiquidity or feed failures — and how you…

→

Adjusted Prices

Adjusted series

Adjusted prices are a historical series restated to remove the mechanical jumps caused by corporate actions, producing a continuous series whose perc…

→

Time Zones & Timestamps

Timestamps

Time-zone and timestamp handling is the discipline of representing when each observation occurred consistently — IST versus UTC, exchange versus syst…

→

Data Validation

Data quality

Data validation is the automated gatekeeping layer that verifies market data against explicit schema, range, continuity and cross-source checks befor…

→

Frequently asked questions

What data do I need for algorithmic trading?

At minimum, historical price data (OHLC bars or ticks) for backtesting and a live data feed for trading, plus reference data on corporate actions (splits, dividends, bonuses) to adjust prices correctly. The granularity — daily, minute or tick — depends on your strategy's holding period. Data quality and correct adjustment matter more than sheer volume.

Why are adjusted prices important?

When a stock splits or pays a dividend, its raw price jumps for a purely mechanical reason, not a real move. Adjusted prices remove these artefacts so that returns are continuous and correct. Backtesting on unadjusted data creates fake gaps that can trigger phantom signals or corrupt indicators — a common and serious data error.

What is the difference between tick data and OHLC data?

Tick data records every individual trade or quote with a timestamp and price/size; OHLC data summarises a time interval into four numbers — open, high, low and close. Tick data is far larger and captures intrabar detail (needed for microstructure or precise fill modelling); OHLC is compact and sufficient for most bar-based strategies.