Applying statistical pattern recognition to commodity trading systems by Scott Brill
Statistical pattern recognition (SPR) is a subfield of artificial intelligence concerned with automatic
recognition of meaningful regularities in noisy or complex environments. Since the early 1960s, a large
body of these techniques have been developed to solve problems ranging from machine vision and
machine recognition of human speech to bankruptcy prediction. In this article, we will introduce you to
some basic SPR tools which can be used to create automatic trading systems, and give two
demonstrations of how these may be applied to a weekly gold trading scenario using a personal computer.