Order flow and the bid-ask spread: An empirical probability model of screen-based trading

A probabilistic framework for the analysis of screen-based trading activity is presented. Probability functions are derived for the stationary distributions of the best bid and offer, conditional on the order flows. By identifying the unobservable order and acceptance flows, our estimation method permits the prediction of the stationary distributions of other market statistics. A test is proposed that allows a comparison of predicted and sample bid-ask spread distributions taking parameter estimation error into account. The methodology is applied to the screen-based interbank foreign exchange market, using continuously recorded quotes on the Deutschemark/US dollar exchange rate.

Duke Scholars

Author Tim Bollerslev Economics