News for hackers. Stuff that matters.
Re:Finance novice beats hedge fund pros, winning $100k in Qu... by jeffreyrogers on Thursday March 05, 2015 @03:09PM (#9153480)
Finding anything meaningful in results like this is a bit tough since everything is looking backwards. It would have been nice if they said (for example) "okay, here's the deadline, you can't make changes to your algorithm after this date, and then we'll test it over the next six months of data."

Also, I wonder whether they're taking costs into account since any strategies that involve making lots of trades will quickly see their profitability drop due to fees.

Re:Finance novice beats hedge fund pros, winning $100k in Qu... by dunster on Thursday March 05, 2015 @03:16PM (#9153515)
It's both forward-testing and backward testing. The algos have been locked since submission - some were submitted as early as 1/15, all were submitted by 2/2. That makes it both an in-sample and out-of-sample test.

Yes, the Quantopian platform includes default commissions. It also includes default slippage. No model is perfect, of course, but this is a tool that's had a lot of development.

(I work at Quantopian)

Re:Finance novice beats hedge fund pros, winning $100k in Qu... by 21echoes on Thursday March 05, 2015 @06:31PM (#9154642)
isn't that effectively how it worked? "a combination of live paper trading results during [the month after submission], as well as two years of backtesting against historical market data"