The key feature that unites all types of high-frequency trading strategies is persistence of the underlying tradable phenomena. This part of the book addresses the ways to identify these persistent trading opportunities.
High-frequency trading opportunities range from microsecond price moves allowing a trader to benefit from market-making trades, to several-minute-long strategies that trade on momentum forecasted by microstructure theories, to several-hour-long market moves surrounding recurring events and deviations from statistical relationships. Dacorogna et al. (2001) emphasize a standard academic approach to model development:
Document observed phenomena.
Develop a model that explains the phenomena.
Test the model's predictive properties.
The development of high-frequency trading strategies begins with identification of recurrent profitable trading opportunities present in high-frequency data. The discourse on what is the most profitable trading frequency often ends once the question of data availability emerges and researchers cannot quantify the returns of strategies run at different frequencies. Traders that possess the data shun the public limelight because they are using the data to successfully run high-frequency strategies. Other sources tend to produce data from questionable strategies.
The profitability of a trading strategy is bound by the chosen trading frequency. At the daily trading frequency, ...