Tick Data Analysis

Overview

Tick data analysis examines individual trades and quotes to understand market dynamics, detect patterns, and develop trading signals. It is the foundation of high-frequency and intraday trading strategies.

Difficulty advanced

Key Tick-Level Metrics

1. Tick Direction

Classify each trade as buyer-initiated or seller-initiated:

Lee-Ready Algorithm:
- Compare trade price to NBBO midpoint
- Trade at/above ask → Buyer-initiated
- Trade at/below bid → Seller-initiated
- Trade at midpoint → Use tick test (compare to previous trade)

Tick Test:
- Trade price > previous trade → Buyer-initiated
- Trade price < previous trade → Seller-initiated

2. Volume Analysis

Tick Volume: Volume of each individual trade
Total Volume: Sum of all ticks in period
Average Tick Size: Total Volume / Number of Trades

Large trades (block trades) signal institutional activity.
Small trades often represent retail or algorithmic activity.

3. Price Discovery

Information share: How much does each venue contribute to price discovery?

Hasbrouck (1995) Information Share:
- Decompose price changes into permanent and transitory components
- Permanent component = information
- Venue with highest permanent share = price leader

Checklist

• [ ] Trade classification method appropriate for data
• [ ] Tick bar type matches strategy (volume/dollar/tick)
• [ ] Time synchronization verified
• [ ] Outlier trades identified and handled
• [ ] Missing data gaps addressed
• [ ] Microstructure noise filtered if needed
• [ ] Order flow metrics calculated
• [ ] Intraday patterns identified
• [ ] Price impact coefficient estimated
• [ ] Volume profile constructed

References

1. Hasbrouck, J. (2007). Empirical Market Microstructure. Oxford University Press.
2. Lee, C. & Ready, M. (1991). "Inferring Trade Direction from Intraday Data." Journal of Finance, 46(2), 733-746.
3. Kyle, A.S. (1985). "Continuous Auctions and Insider Trading." Econometrica, 53(6), 1315-1335.