Machines have already proven their ability to recognize an image of a cat from an image of your morning cup of coffee, for example, but in the low-signal-to-noise environment of finance, machines have not yet been so discerning. In this short video from AQR, learn why our industry still struggles to answer the question, “Can machines learn finance?” even while putting ML to good use in their investment processes.
Dear Investment Professional,
Not everyone has to be a master coder or a world-class quant, but it’s becoming increasingly important for investors of all stripes to understand how data science, artificial intelligence (AI), and machine learning (ML) will impact investment processes. Across asset classes, more managers are utilizing these tools to generate differentiated sources of return, but allocators also see the incremental benefits of the operational efficiencies and scale that AI and ML have introduced. On the flip side, as I wrote in a recent What About Beta? post1, allocators need to be wary of overfitting and “machine-washing” in this space. Indeed, everyone seems to claim that they incorporate machine learning in their techniques, but as definitions keep evolving, responsible investors should be able to discern when these technologies are simply automating DCF spreadsheets to save time or offering access to genuine sources of alpha.
The CAIA Content Team
John L. Bowman, CFA, Senior Managing Director, Twitter, LinkedIn
Keith Black, PhD, CFA, CAIA, FDP, Managing Director Content Strategy, Twitter, LinkedIn
Guowei Jack Wu, CFA, CAIA, Content Director, APAC, Twitter, LinkedIn
Aaron Filbeck, CFA, CAIA, CIPM, FDP, Director, Global Content Development, Twitter, LinkedIn
