Abstract
This article discusses simulations of maker loss in a limit-order market. The author assumes that if price moves in the adverse direction before the maker order is executed, the order is canceled and resubmitted at a new best price. The author’s goal is to check if such a strategy may lead to a loss exceeding the taker loss (the bid/offer spread). We describe three models that differ in the level of explicit usage of market data. For the model calibration, we use historical market data from the inter-bank FX spot brokerage ICAP/EBS. In the first model, price dynamics are simulated with a random trinomial tree function, while the initial order book size and its depletion rate are simulated using probability distributions drawn from the EBS market data. In the second model, real EUR/USD rates rather than simulated prices are used. In the third model, both prices and initial order book size are taken from the EBS market data records; only the order book depletion rate (which is not an observable variable) is simulated. All three models yield an expected maker loss well below one pip, which points at an advantage of the maker trading strategy over the taker one. <b>TOPICS:</b>Simulations, statistical methods, global
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have