Stochastic Control Theory in High-Frequency Trading


Douglas Borden, Knight Equity Markets 


Trading decisions in High Frequency Trading involve a subtle interplay between expected price movement, transaction costs, market impact and risk. And these decisions need to be made in milliseconds, millions of times a day. Standard practice is to derive a set of trading rules, with the rule parameters optimized through simulation backtesting. In this talk I present a different approach to making High Frequency Trading decisions, making use of techniques from Stochastic Control Theory. I derive the basic equations governing high-frequency decision making and describes techniques for solving the resultant equations.


Please click on this link to view the presentation.