Hacker News Reader

Comment by nicholas73 | Hacker News Reader

Quantopian hosts one where the winner gets actual money funded for their algorithm.

Thank you. I have used Quantopian but they are geared more toward a day-trader type of mentality:

https://www.quantopian.com/allocation 'We are looking for algorithms that turn over their portfolio between 12 and 500 times per year (e.g between once per month and twice a day).'

I hold my stock picks for one year, so Quantopian definitely doesn't work for me.

Would you say your long holding period should make you less confident about beating the market long term? Even though you tested from 2001, the actual number of "samples" (end-to-end trades) is actually very low. Maybe that's why Quantopian wants algorithms with higher turnover.

Hi, I'm the founder of Q. You're right about the turnover, though it is a multi-faceted trade-off. While increasing the trading frequency accelerates the accumulation of data, it also increases drag from transaction costs and that tends to lower a strategy's capacity.

Here is a post and video that talks about many of the criteria we use to evaluate algorithms: https://www.quantopian.com/posts/how-to-get-an-allocation-wr...

I don't have a great source link for you (forgot) but you only need a very small number of companies to roughly match a given index: 20-ish. I'm saying my 25 companies will handily beat the S&P 500 over the long run. I'm trading with real money in a public forum so time will tell.

Reply on news.ycombinator.com