cryptogon.com

Quantitative strategists, or “quants” as they are known, attempt to profit from pricing inefficiencies identified through mathematical models. These send buy and sell signals on small variations in price between different securities.

One hedge funds manager said the average quantitative fund manager was down about 15 per cent in the first few days of August.

“Nothing seems to be working. Previously uncorrelated factors have recently been falling with the same pace, leaving investors with very few places to hide,” said Citigroup analysts in a report to clients last week.

Hedge fund managers who have suffered in the first few days of August and late July include James Simons, long acknowledged as the “king of quants”, at Renaissance Technologies and Clifford Asness of AQR Capital Management. Quantitative managers at Goldman Sachs and Highbridge Capital Management have also experienced difficulties.

Statistical arbitrage funds have run into particular problems. Their mathematical models rely on past trading patterns to predict how particular securities will perform in the future if other securities, say, fall in price. But their models are unlikely to take into account current trading conditions where investors, desperate to raise cash, are selling everything they can.

They are also unwinding short positions, which means buying back stock they sold earlier. Thus, companies with poor prospects have seen shares rise and vice versa, undermining the logic of “stat arb” models. Compounding the problem is that many stat arb managers have borrowed heavily to buy shares.

One hedge fund manager estimated that statistical arbitrage funds with more than $100bn (€73bn, £49bn) in assets had on average borrowed four times their actual assets. These borrowings magnify significantly any moves they may make in the market.

It is these more heavily indebted statistical arbitrage funds that have proved most attractive for pension funds seeking supposedly lower risk hedge fund strategies.