https://www.ft.com/content/0103452e-28f9-11e7-bc4b-5528796fe35c Man Group has appointed an industry veteran to head machine learning for GLG, its discretionary trading unit, as hedge funds bet that the future is in computer algorithms and big data. William Ferreira, who previously worked at Man’s AHL unit, which trades using computer algorithms, rejoined the company in the new role this month after a stint at another quant fund. Discretionary hedge fund managers, especially those who trade equities, are increasingly turning to quantitative analysis to help them predict market moves, a technique that had once been left to the systematic trading funds. It has been dubbed “quantamental” trading by some. “Machine learning” is a step beyond quant trading. A branch of artificial intelligence, it is a sophisticated computing technique that allows hedge funds to find patterns in data using algorithms that can improve and learn over time as the data sets expand. BlackRock and State Street are among asset managers investing in machine learning, while the giants of hedge fund quantitative trading — Renaissance Technologies, DE Shaw and Two Sigma — have experimented with the technique for years. Man GLG is one of the first discretionary funds to have established a head of machine learning. We see many opportunities to utilise machine learning across the diverse data sets available to the discretionary investment business WILLIAM FERREIRA “We are continually seeking to develop our offering for our clients and, as the amount of data available continues to expand, these techniques can supplement existing rigorous quantitative and qualitative analysis,” said Teun Johnston, chief executive of Man GLG. The amount of data available for hedge funds to utilise has exploded in the past few years, from credit card data that can give traders an insight into where people are spending their money ahead of companies announcing their earnings, to analysing GPS tracking or satellite surveillance to determine whether foot traffic in stores is up or down. Some funds also use web-scraping tools, which for example can show which products are suddenly being offered at a discount, an indication that sales could be lower than expected. Using machine learning at GLG also allows Man to capitalise on clients who still want most of their money managed by humans, while still using the same techniques to analyse the ever-expanding amounts of data available. “We see many opportunities to utilise machine learning across the diverse data sets available to the discretionary investment business,” Mr Ferreira said. He will work with the fund’s traders on using machine learning to interpret news and social media, market events and announcements, and visualising complex data. Mr Ferreira most recently worked at Florin Court Capital, a systematic fund run by Doug Greenig, a former chief risk officer of AHL. Man GLG’s latest hire also spent time at the hedge funds GSA and CQS, and at JPMorgan and Merrill Lynch. Man Group, the world’s largest publicly traded hedge fund, has been investing in developing its machine learning techniques for several years through its AHL unit. Last year, it announced that the partnership it funds with Oxford university, the Oxford-Man Institute, would expand its focus to machine learning and data analytics.