The ability to postpone one's execution in the market without penalty in search of a better price is an important strategic advantage in high-frequency trading. To elucidate competition between traders one has to formulate to a quantitative theory of formation of the execution price from market expectations and quotes. Equilibrium theory was provided in 2005 by Foucault, Kadan and Kandel. I derive an asymptotic distribution of the bids/offers as a function of the ratio of patient and impatient traders using the dynamic version of the Foucault, Kadan and Kandel limit order book (LOB) model. Our version of the LOB model allows stylized, but sufficiently realistic representation of the trading markets. In particular, dynamic LOB allows simulation of the distribution of execution times and spreads from high-frequency quotes. Significant analytic progress is made toward framing of short-term trading as competition for immediacy of execution between traders under imperfect information. The results are qualitatively compared with empirical volume-at-price distribution of highly liquid stocks.
Benchmark dataset for mid-price forecasting of limit order book data with machine learning methods
