What can AI do for investment portfolios? A case study

Strategies based on artificial intelligence (AI) are increasingly applied in investing and portfolio management. Their contexts, usefulness, and results vary widely, as well as its ethical implications. Yet for a technology that many anticipate will transform investment management, AI remains a black box for far too many investment professionals.

To shed some light on the subject, we zeroed in on a particular AI stock trading model and explored what it can bring in terms of risk-related benefits and costs. Use of proprietary data provided by Trader AIan AI business model led by our colleague Ashok Margam and team, we look at their decisions and overall performance from 2019 to 2022.

Trader AI has few restrictions on the market positions it takes: it can go both long and short and change positions at any time of the day. However, at the close of each day, you exit the market completely, so your positions are not held overnight.

So how did the strategy fare in different time frames, trading patterns, and volatility environments? And what can this tell us about how AI could be applied more broadly in investment management?

Trader AI outperformed its benchmark, the S&P 500, over the three-year analysis period. While the strategy was long versus short neutral, its beta on the time frame was statistically zero.

Traders AI Model vs. S&P 500 Monthly Equity Curve ($10k Investment)

Trader AI took advantage of moments of greatest asymmetry to achieve these results. While the S&P 500 had a negative skew or strong tail to the left, the AI ​​model showed the opposite: skew to the right or a strong tail to the right, which means that the trader AI had few days in which generated very high yields.

So where was the model most successful? Was it better to go long or short? On days of high or low volatility? Do you choose the right days to stay out of the market?

About the last question, the traders AI actually avoided trading on high performance days. You can anticipate high-risk premium events and choose not to take a position on the direction the market will take.

Trader AI performed better on a market-adjusted basis when going short. It gained 0.13% on average in its short days, while the market lost 0.52%. Therefore, the model predicted low days better than high days. This pattern is also reflected in bear markets, where trader AI outperformed relative to bull markets.

Finally, the AI ​​model performed better on high volatility days, outperforming the S&P 500 by 0.19% per day on average, while underperforming on low volatility days.

AI Model Percentage Return vs. VIX Percentage Change

In general, the Traders AI results demonstrate how a particular AI stock trading model can work. Of course, it hardly serves as a proxy for AI applications in investing in general. However, the fact that it was better at predicting down days than up days, that it was successful when volatility was high, and that it avoided trading all together before big market-moving events are critical data points. . In fact, they hint at the enormous potential of AI to transform investment management.

For more on this topic, don’t miss “Ethics and Artificial Intelligence in Investment Management: A Framework for Practitioners”, by Rhodri Preece, CFA.

If you liked this post, don’t forget to subscribe to Entrepreneurial Investor.

All messages are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Image credit: ©Getty Images/Svetlozar Hristov

Professional Learning for CFA Institute Members

CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content in Entrepreneurial Investor. Members can easily register credits using their online PL tracker.