Can We Learn to Beat the Best Stock

TL;DR

This paper introduces a novel stock trading algorithm that leverages statistical relations between stock pairs, demonstrating its ability to outperform the market and the best individual stocks through empirical analysis.

Contribution

It presents a new approach based on statistical relations between stocks and a smoothing technique for expert learning parameters, outperforming traditional methods.

Findings

Algorithm beats the market in historical data

Outperforms the best stock in the market

Uses a new smoothing technique for parameters

Abstract

A novel algorithm for actively trading stocks is presented. While traditional expert advice and "universal" algorithms (as well as standard technical trading heuristics) attempt to predict winners or trends, our approach relies on predictable statistical relations between all pairs of stocks in the market. Our empirical results on historical markets provide strong evidence that this type of technical trading can "beat the market" and moreover, can beat the best stock in the market. In doing so we utilize a new idea for smoothing critical parameters in the context of expert learning.