ABSTRACT
This chapter outlines principles and ideas that are probably more relevant than the sum of technical details covered subsequently. The blossoming of machine learning in factor investing has it source at the confluence of three favorable developments: data availability, computational capacity, and economic groundings. First, the data. Nowadays, classical providers, such as Bloomberg and Reuters have seen their playing field invaded by niche players and aggregation platforms. Second, computational power, both through hardware and software. Storage and processing speed are not technical hurdles anymore and models can even be run on the cloud thanks to services hosted by major actors such as Amazon, Microsoft, IBM and Google and by smaller players. Finally, economic framing. Machine learning applications in finance were initially introduced by computer scientists and information system experts and exploited shortly after by academics in financial economics, and hedge funds.
