Machine Learning for Algorithmic Trading and Trade Schedule Optimization

Abstract

In this paper we present a machine learning technique that can be used in conjunction with multi-period trade schedule optimization used in program trading. The technique is based on an artificial neural network (ANN) model that determines a better starting solution for the non-linear optimization routine. This technique provides calculation time improvements that are 30% faster for small baskets (n = 10 stocks), 50% faster for baskets of (n = 100 stocks) and up to 70% faster for large baskets (n ≥ 300 stocks). Unlike many of the industry approaches that use heuristics and numerical approximation, our machine learning approach solves for the exact problem and provides a dramatic improvement in calculation time.