I can't get good results when I can only trade a stock that outperformed 159 stocks instead of 1590.

Have you actually verified that or are you just not getting good results with 160 stocks?

You can switch to daily execution and use 1600 but that's just for backtesting, I guess it won't work in paper trading or live trading.

Hi Jeff,

In the Quantopian research platform, I started to play around with code to screen all approx. 20,000 securities in the Q database. Seems doable, but it takes some time. Presently, the code I wrote is broken (due to a change by the Q engineers), but presumably it'll be fixed. Then, I can imagine doing your screening using zipline as a backtester, within the research platform. My understanding is that eventually, there may be a way to "call" the online backtester from research platform, but I don't think it has been implemented yet.

My sense is that Q hasn't worked out yet how to enable online computationally intensive work, that would feed into a live trading algo. I gather it could be done overnight, using your own data, and then feeding the result in via fetcher. I could imagine doing the same thing on the Q research platform, but having access to the Q databases. With some cleverness, you might be able to pull it off manually, but it is not obvious how one would automate it, since I don't think there is a way to write out data to a local file per a schedule.

Grant

I know about the lesser restriction for daily trading, but as you said Andrea that won't even work in the Open. If I can't backtest my strategies in minute mode then I can't use them.

I just hope Quantopian plans to change this