Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
First post, K filter across as many stocks as you want

Here is an algo I modified from another post. Instead of trading a pair, you can specify a list of stocks. If anyone has an idea on how to optimize I would love to hear suggestions.

5 responses

If the algo is as good as it seems from the performance graph (comparing to benchmark SPY), why not also short when appropriate? Or at least find the metric to "stand asside".

What was the other post with the pair example?

Thanks for the nice illustration.

I hit send too soon - what I mean about that graph is how nicely your performance line matches/mirrors spy. It seems like it is coming close to an ML algo with decent "predictive" qualities.

That's basically what I'm trying to figure out. Also I might be using too wide of a range of stocks. I think it diminishes the predictive power too much. I might try different indices to try and dial in the performance some

Can somebody fix schedule_function in this promising algo?

Here is the same logic across the dow stocks