In two of our previous articles, we explored the idea of using artificial intelligence to backtest trading strategies. Since then, AI has continued to develop, with tools like ChatGPT evolving from simple Q&A assistants into more complex tools that may aid in developing and testing investment strategies—at least, according to some of the more optimistic voices in the field. Over a year has passed since our first experiments, and with all the current hype around the usefulness of large language models (LLMs), we believe it’s the right time to critically revisit this topic. Therefore, our goal is to evaluate how well today’s AI models can perform as quasi-junior quantitative analysts—highlighting not only the promising use cases but also the limitations that still remain.

This is the moment we all have been waiting for, and today, we would like to acknowledge the accomplishments of the researchers behind innovative studies in quantitative trading. So, what do the top five look like, and what will the authors of the papers receive?

Let’s find out …

Quantpedia Days 2025

– Celebrate with us the relentless pursuit of knowledge and ingenuity
– You can now subscribe to any of our services, be it 3-, 12- or 36-months Quantpedia Prime, Premium, or Pro subscription, and get the same 2nd subscription for your co-worker or fellow researcher for free – an offer valid between 23rd May and 1st June 2025
– The winner of the 2nd season of our Quantpedia Awards competition will be announced on Tuesday, 27th May 2025
– What’s your favorite paper from the presented top 10?

Building machine learning models for trading is full of nuances, and one important but often overlooked question is: what exactly should we try to predict—the direction of the next market move or the actual value of the asset’s return? A recent paper by Cheng, Shang, and Zhao, titled “Direction is More Important than Speed” offers a clear and practical answer. Their research shows that focusing on direction—simply whether returns will be positive or negative—leads to better model accuracy and, more importantly, stronger real-world investment performance. This is especially true when using machine learning methods, where predicting the direction allows models to better capture downside risks and build more effective trading strategies. For anyone using ML in finance, this paper makes a strong case that predicting where the market is headed is often more valuable than predicting how far it will go.

What can we realistically expect from investing across different asset classes over the long run? That’s the kind of big-picture question the “Long-Run Asset Returns” paper tackles—offering a sweeping look at how stocks, bonds, real estate, and commodities have performed over the past 200 years. The paper goes beyond just listing historical returns—it explains how reliable (or not) those numbers are by digging into the quirks and issues hidden in very old data. The authors look at what happens to returns when you include countries or time periods that usually get left out, and they show that the past isn’t always as rosy—or as repeatable—as it might seem if you only look at recent decades.

Can machine learning models better predict stock returns if they are tailored to specific industries, or is a one-size-fits-all (generalist) approach sufficient? This question lies at the heart of a recent research paper by Matthias Hanauer, Amar Soebhag, Marc Stam, and Tobias Hoogteijling. Their findings suggest that the optimal solution lies somewhere in between: a “Hybrid” machine learning model that is aware of industry structures but still trained on the full cross-section of stocks offers the best performance.