Quant Revolution

Welcome, officially, to the new version of the newsletter! Add it to your primary inbox and make sure you aren’t getting delivery issues. As a reminder, I read every response personally and it goes directly to my inbox. Feel free to share thoughts, feedback, suggestions, ideas, or funny memes.

This week, I want to dig into quant hedge funds, a brief history of their rise, and what emerging AI technology may do to these funds in the future. It’s exciting times for these funds and for financial modeling in general, and AI will enhance their ability to process massive swaths of data to further enhance their algorithms.

Jim Simons

We can’t discuss quant funds without giving a bit of back story on the man considered by many to be the father of quant investing. Jim Simons was a math professor and codebreaker who contributed research to pattern recognition and modern string theory. He used these mathematical concepts to found Renaissance Technologies in 1982. The returns of Renaissance, and in particular its Medallion Fund, earned Simons the title of “greatest hedge fund manager of all time” prior to his passing this year.

The core of Renaissance’s success is that they built their investment strategies around huge volumes of data that only they had access to. By purchasing extensive data sets, they were able to build secretive trading algorithms that other funds could not mirror. Renaissance analyzed and processed far more data than any rival fund while automating the majority of their trades to remove the possibility of human error.

Quant Investing

So what, exactly is quant investing? How does a so-called “quant fund” differ from traditional hedge funds?

1. Data-Driven: Quant funds utilize mathematical models, algorithms, and vast amounts of data to inform investment decisions. Traditional hedge funds rely on fundamental analysis of businesses and broader economic data.

2. Automation: They rely heavily on automated systems to execute trades based on predefined criteria instead of relying on human judgement.

3. Talent Pool: They employ scientists, mathematicians, and engineers rather than traditional finance professionals.

4. Strategy: They focus on identifying and exploiting statistical patterns and anomalies in the market instead of analyzing specific companies. Alternatively, traditional hedge funds employ a wide range of strategies such as long-short equity, global macro, event-driven, and other strategies based on market conditions and forecasts.

They have risen in popularity due, in large part, to their overwhelming success compared with other methods. The majority of hedge funds lose money when compared to a standard market benchmark. This is often due to errors in human judgement, adherence to specific strategies, and the pure fact that predicting the market is impossible.

Taking a data-driven approach and using market indicators to inform decisions has proven to be a far more effective approach in many instances. Additionally, computing power has made these technologies increasingly powerful over the years. Combing through the amount of data that Renaissance levereged in its strategies takes a tremendous amount of computing horsepower that was not previously possible.

In fact, another famous hedge fund investor, Ray Dalio, discusses often how his fund, Bridgewater, benefitted immensely from advances in technology. His models used to be done by hand, and it was a painstaking (and slow) process. Being able to plug these formulas into a computer, and run everything at once, made their investment strategies faster, more reliable, and more efficient.

So what’s next?

We are now entering the era where large language models (LLM’s) like ChatGPT have access to, and have been trained on, massive data sets. One of the strengths of these LLM’s is their ability to comb through large quantities of data and create viable models from that data.

We now have technology capable of creating statistically more accurate financial models on a consistent basis. While these models still need human oversight and intervention, they have proven to be better and more reliable than previous, man-made models.

Given that the underlying algorithms powering quant funds will only be made more powerful by the ability of LLM’s to process and model data more effectively, I think we will see these funds boast even greater returns in the near future. They are essentially receiving exponentially more computing power as these language models evolve. The same technological fuel that helped stoke the fire of the initial quant revolution will now allow it to evolve into an even better version of itself.

Credibility and the future of finance

Fear not if you currently manage a large book and are worried about new-age quant funds taking your business. The financial industry is built on credibility and reputation above all else. Plenty of funds get sub-par returns, but they are led by upstanding people with good reputations who manage risk well. It’s a relationship driven industry because you have to fully trust the person you invest your money with.

The established quant funds will get better thanks to the rise of AI. But many people are still wary of this technology, and new funds trying to leverage it to woo clients will find themselves needing to overcome the large reputational hurdle that comes with initial investments. Even though new technology will empower smaller, start-up funds to generate impressive returns, they will still struggle to raise capital unless they have an industry pedigree or get someone reputable to vouch for them.

The flip side of this is that established funds who are looking to evolve may leverage a new talent pool to do so. Quant-focused funds already look to math and sciences for new hires. For those out there with an interest in finance who happen to be science-minded, spending time learning how LLM’s work is a worthwhile investment.

For that matter, anyone looking to level up their careers in the future should do everything they can to get in front of this technology in some capacity. Take job opportunities with people building in the space. Spend time reading, researching, and writing about AI. Leverage these tools in your own workflows to get more comfortable. AI will accelerate our workforce, and those who get ahead of the curve are poised to benefit most from this acceleration.

What I’m interested in this week

1. The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution by Gregory Zuckerman

2. “Squarespace to Go Private in $6.9B All-Cash Transaction with Permira” — this is a 29% premium for shareholders over the 90-day average trading price

3. “Why the Medallion Fund is the Greatest Money-Making Machine of All Time” by Nick Maggiulli

4. “AI and quantitative investing: the beginning of a beautiful friendship?” by David Wright (no relation, I promise)