Having said that, there is plenty of interesting papers in this area, and from the abstract, this paper seems to take an interesting approach too.
Having said that, there is plenty of interesting papers in this area, and from the abstract, this paper seems to take an interesting approach too.
http://news.morningstar.com/articlenet/article.aspx?id=64857...
Of course, contra your ML researcher, this presupposes that there are effective strategies that are published.
Actually, I think the bigger point is that researcher's rarely "tell all." Just like Moriarty in the final episode of Season 2 of Sherlock (on BBC 2013-2014), sometimes a work is created as an advertisement demonstrating the ability to draw larger returns. Those intimate with the data will always have more than than will or can possibly publish.
The Fama-French 3 factor model still holds. (http://en.wikipedia.org/wiki/Fama%E2%80%93French_three-facto...) Part of this is because Gene Fama doesn't have a hedge fund on the side. He advises DFA, but they want their customers to understand the model. (He also won this year's Economics Nobel)