Abstract—In this paper, we propose a Bayesian inference of
the Markov chain model class to model dynamics of order book
in high frequency trading environment. Accordingly, software
program can predict the move of market price for both ask &
bid via predictive distribution. A strategy algorithm can be
developed for generating, routing & executing orders to gain
profit. Experimental result based on security AAPL showed
over 98% coverage by 50 transitions from 6561 state space. It
further indicated market behavior of short time-frame can be
clustered & labeled.
Index Terms—Inferring markov chain, bayesian inference,
high frequency trading, order book.
Yuan Lung Chang is with New York, NY10128 USA (e-mail:
jeffrey.chang.dc@gmail.com)
Cite: Yuan Lung Chang, "Inferring Markov Chain for Modeling Order Book Dynamics in High Frequency Environment," International Journal of Machine Learning and Computing vol. 5, no. 3, pp. 247-251, 2015.