We develop metrics based on Shapley values for interpreting time-series forecasting models, including "black-box" models from machine learning. Our metrics are model agnostic, so that they are applicable to any model (linear or nonlinear, parametric or nonparametric). Two of the metrics,...

This article reviews the history of the recent shift to electronic trading in equity, foreign ex- change and fixed-income markets. We analyze a new data set: the eSpeed (Cantor Fitzgerald) electronic Treasury network. We contrast the market microstructure of eSpeed with the tradi- tional voice...