Evaluating the performance of adapting trading strategies with different memory lengths

TL;DR

This paper introduces a prediction model for trading strategies based on the minority game, which evaluates multiple memory lengths to improve prediction accuracy and investment returns in stock trading.

Contribution

The paper presents a novel prediction model that adaptively evaluates trading strategies with different memory lengths, demonstrating improved success rates and returns over traditional strategies.

Findings

Prediction success rate over 51.5% on S&P 500 stocks

Higher returns than buy-and-hold strategy

Predictions remain profitable after trading costs

Abstract

We propose a prediction model based on the minority game in which traders continuously evaluate a complete set of trading strategies with different memory lengths using the strategies' past performance. Based on the chosen trading strategy they determine their prediction of the movement for the following time period of a single asset. We find empirically using stocks from the S&P500 that our prediction model yields a high success rate of over 51.5% and produces higher returns than a buy-and-hold strategy. Even when taking into account trading costs we find that using the predictions will generate superior investment portfolios.