Hedge funds that use AI to help with trading have been outperforming the hedge fund benchmark for the past year. Preqin Pro tracks performance information for 152 artificial intelligence (AI) hedge funds. Based on three-year cumulative returns, AI hedge funds have outperformed the Preqin All-Strategies Hedge Fund benchmark by a margin of three percentage points, with AI funds returning +26.96% over the past three years and all hedge funds returning +23.87%, as seen in the chart above.
In isolation, this difference does not seem significant enough to give AI funds an edge in the market, but other risk metrics must be taken into account.
Parsing the three-year volatility and Sharpe ratio data, AI funds have slightly more favourable risk-adjusted parameters than the Preqin All-Strategies Hedge Fund benchmark. Over a three-year horizon, AI funds have shown 3.20% volatility and a Sharpe ratio of 1.96, while all hedge funds have posted volatility of 3.87% and a Sharpe ratio of 1.40. The greater a portfolio's Sharpe ratio, the better its risk-adjusted performance. The higher Sharpe ratio, coupled with the lower volatility over a three-year period, suggests that funds utilizing AI technology have indeed been the better investment over the past three years.
AI as a Tool
AI/machine learning is an essential tool for many hedge fund firms, especially quantitative hedge funds (where securities are selected by algorithm or systematic programs) such as Man AHL, Bridgewater, Citadel and D.E. Shaw. These firms use quantitative models to develop new trading strategies, as well as identify new themes and trading signals. The ‘quants’ input these themes and signals into trading systems. But since they often require reprogramming by human quantitative analysts, they are considered ‘pre-AI’ models. This is where the ‘pure’ AI models, self-sufficient models that don’t require human programming, come into the picture. These models, unlike pre-AI quants, adapt to changing market conditions with significantly greater autonomy as they don’t rely on a human component. Some recently launched firms that incorporate this refined version of AI modelling include Cerebellum Capital, Taaffeite Capital Management and Numerai.
Why Are AI Funds Outperforming Humans?
AI funds have a critical advantage over human-run funds: time. Since they process massive amounts of data quickly, they can adapt to rapidly changing market conditions. Machine learning also means the AI model can automatically update itself as it gathers new data, with no need for human oversight – this could be an especially pertinent benefit in an uncertain market.
Many of these AI systems dig through various social media venues to gauge consumer, market and investor sentiment on a particular asset or security. The managers may then use this data to create a proprietary indicator, on which they will base their trading decisions. For example, if an AI model were to determine that market participants felt bearish about the S&P 500 Index, the manager may decide to take a contrarian long position on the market.
Many hedge funds are secretive about the exact strategy they pursue. When an AI fund analyzes the myriad of data pools at its disposal, it can rapidly make connections and locate links between data points that would not be clear to a traditional fund manager. Machine learning allows funds to automatically update as they digest and interpret data, ensuring they become smarter and more efficient as time goes on.
Given the advantages afforded by this cutting-edge technology, it is no surprise to see AI hedge funds gaining a competitive edge in the market and outperforming their human counterparts.
Launching into the Future with AI
As AI/machine learning capabilities evolve and become more sophisticated, more hedge funds are using AI trading methods. The number of AI hedge funds launched in 2018 was up 77% on 2016. In 2012, there were just 12 AI hedge funds launched; 2019 YTD has already matched that number. Although the universe of AI hedge funds is currently comparatively small – Preqin Pro currently tracks 305 such funds, representing aggregate assets under management of $17bn – we expect it to grow. While the overall metrics are not significantly different, the data shows a statistical performance advantage for AI hedge funds over hedge funds as a whole.
Looking for more hedge fund performance data? Preqin Pro offers fund-by-fund and industry-level performance data, as well as fully customizable and transparent benchmarks by strategy and geographic focus. Take a demo today to find out how Preqin Pro can help you.