tickdata

Tick data refers to time-stamped records of every price change and order book event in financial markets, capturing trades and quote updates at the finest granularity available. The term encompasses both trade ticks and quote ticks, and is often described in terms of the depth of data, such as Level I or Level II.

A tick record typically includes a timestamp, price, and size for trades; for quotes, bid and ask prices and sizes, and sometimes the exchange or data provider. Some datasets separate trades and quotes, while others interleave them in a single stream or storage structure. Tick data can be delivered as a live feed or stored in historical databases for later analysis.

Tick data differs from bar data, which aggregates information into fixed intervals like seconds, minutes, or days. Tick data preserves intrabar price movements and order book changes, enabling detailed market microstructure analysis and more precise model testing, but it comes with higher storage and processing costs.

Common uses include backtesting trading strategies, researching price formation and liquidity, estimating volatility at high frequency, and calibrating high-speed trading models. Researchers and traders rely on tick data to study how prices respond to order flow and to measure market depth and resiliency.

Storage and formats vary; tick data is vast and is stored in flat files (CSV, binary formats) or in databases, with timestamps at microseconds or nanoseconds. Normalization of symbols, venues, and time zones is important, and data is often compressed and indexed to support fast querying.

Quality considerations are central to tick data work. Cleaning needed to address duplicates, outliers, erroneous ticks, missing values, and corporate actions is common, as is alignment across venues when comparing across markets or instruments. Tick data is typically obtained from exchanges or specialized data vendors under licensing terms.