Academic shines light on data mining in alternative beta

Sharpe ratios on complex products fall 73% compared with backtests

A recent study has highlighted over-fitting in alternative beta products, though the situation is improving

If systematic investing has one obvious weakness it is the problem of over-fitting – shaping strategies too closely to fit historical data, and so dooming them to underperform when put to work in reality.

There have been multiple studies to show over-fitting exists in the industry, but until recently the precise scale of the problem has been hard to pin down.

Academic Antti Suhonen – a professor of finance at Aalto University in Finland and former banker – has sought to do just that. His