Abstract
The numerical nature of financial markets makes market forecasting and portfolio construction a good use case for machine learning (ML), a branch of artificial intelligence (AI). Over the past two decades, a number of academics worldwide (mostly from the field of computer science) produced a sizeable body of experimental research. Many publications claim highly accurate forecasts or highly profitable investment strategies. At the same time, the picture of real-world AI-driven investments is ambiguous and conspicuously lacking in high-profile success cases (while it is not lacking in high-profile failures). We conducted a literature review of 27 academic experiments spanning over two decades and contrasted them with real-life examples of machine learning-driven funds to try to explain this apparent contradiction. The specific contributions our article will make are as follows: (1) A comprehensive, thematic review (quantitative and qualitative) of multiple academic experiments from the investment management perspective. (2) A critical evaluation of running multiple versions of the same models in parallel and disclosing the best-performing ones only (“cherry-picking”). (3) Recommendations on how to approach future experiments so that their outcomes are unambiguously measurable and useful for the investment industry. (4) An in-depth comparison of real-life cases of ML-driven funds versus academic experiments. We will discuss whether present-day ML algorithms could make feasible and profitable investments in the equity markets.
Highlights
This article will analyze 27 peer-reviewed articles describing experiments in artificial intelligence (AI) market forecasting over the past two decades
In order to relate academic experiments to market practice, our case studies focus on funds where machine learning is used in the investment decision-making process
The second most popular unit of forecasting accuracy was root mean square deviation (RMSD), otherwise known as root mean square error (RMSE), which was used in 6 articles
Summary
This article will analyze 27 peer-reviewed articles describing experiments in AI market forecasting over the past two decades (the details of the inclusion criteria are in Appendix A). Most of them focus on forecasting an entire market (proxied by a benchmark equity index). We will analyze the existing market data on ML-driven investment vehicles (“AI funds”). – Market forecast—articles where ML algorithms attempted to predict the performance of one or more selected markets, proxied by a benchmark index. In this setup the focus is forecast and the ability to generate trading signals (buy, sell, hold, short). – Bespoke portfolio construction—articles where AI algorithms attempted to predict the performance of a number of equities and build a profitable portfolio, autonomously determining asset allocations (weights). In this remainder of this section we will introduce a number of considerations and clarifications relevant to the use of AI in investment decision-making in general and to our article in particular
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Data Science and Analytics
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
