Chapter 9. Working with Tick Data

Trading opportunities are largely a function of the data that identifies them. As discussed in Chapter 7, the higher the data frequency, the more arbitrage opportunities appear. When researching profitable opportunities, therefore, it is important to use data that is as granular as possible. Recent microstructure research and advances in econometric modeling have facilitated a common understanding of the unique characteristics of tick data. In contrast to traditional low-frequency regularly spaced data, tick data is irregularly spaced with quotes arriving randomly at very short time intervals. The observed irregularities present researchers and traders with a wealth of information not available in low-frequency data sets. Inter-trade durations may signal changes in market volatility, liquidity, and other variables, as discussed further along in this chapter.

In addition, the sheer volume of data allows researchers to produce statistically precise inferences. As noted by Dacorogna et al. (2001), large sets of data can support considerably wider ranges of input variables (parameters) because of the expanded number of allowable degrees of freedom.

Finally, the copious quantities of tick data allow researchers to use short-term data samples to make statistically significant inferences pertaining to the latest changes in the markets. Whereas a monthly set of daily data is normally deemed too short a sample to make statistically viable predictions, volumes ...

Get High-Frequency Trading: A Practical Guide to Algorithmic Strategies and Trading Systems now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.