Beyond ChatGPT: Quant Investing in the New Age

Quantitative asset allocation in the age of AI

As Artificial Intelligence becomes more prevalent in our economy, allocators are looking how to harness it for investment purposes. Today, ChatGPT is the most popular such AI. As the sector advances though, technology will get better, and allocators will need to keep up. Below investors can find some of the latest research on how AI can help the allocation process.

Harnessing GPT For Smarter Asset Management: Prospects And Perils (Robeco)

GPT falls into a general class of machine learning models called generative models. Learn more about how these quant tools are impacting asset allocation processes and decisions.

Dynamic Multifactor Strategies – A Macro Regime Approach (Invesco)

For compliance reasons, this paper is only accessible in the United States

Periods like the late 1990s and 2010s are reminders that factor investing often experiences multi-year periods of dramatic deviations from expected returns.

Can ChatGPT Help Investors Process Information? (Chicago Booth)

The GPT-3.5 model, which has been the foundation for ChatGPT, is particularly well-suited for analysing corporate disclosures due to its ability to summarise relevant information.

Institutional Investors’ Impact on Factor Premiums (Alpha Architect)

Over time markets are becoming more and more efficient. As they do so, the hurdles to active managers generating alpha increase. Here is where quant tools step in.

Can We Backtest Asset Allocation Trading Strategy in ChatGPT? (Quantpedia)

This in-depth piece of analysis looks at whether ChatGPT can really become a fundamental tool for asset allocators or not.

Low Volatility, The Hidden Factor (BNP Paribas AM)

At the portfolio level, minimum variance, maximum diversification and equal-risk contribution strategies all eventually rely on the low volatility anomaly as a source of alpha.

Predictions, Profits, and Delays in Prices

The authors argue that return predictability arises because of delays in prices. Is this the reason why paper profits often do not translate into real-world profits?

Using Machine Learning to Adapt Capital Allocations to Market Regimes

The authors trained a neural network to predict the future 1-month Sharpe ratio based on any hypothetical set of trading parameters and the current market features.

Can Factors Help Understand the Risks Associated with U.S. Banks? (Two Sigma)

If banks continue to make headlines, this paper argues that factor analysis may provide important historical context of their risk exposure.

Applying ML Portfolio Modeling to Bitcoin (Fidelity Digital Assets)

The purpose of the model used in this paper is to inform investment decisions by simulating market environments to provide a range of different outcomes.