Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Generating the Bayesian Forecast Cone Graph in Research (plus a bonus!)

If you've used pyfolio tear sheets then you've already seen the cumulative return graph with a forecast cone. Unfortunately, the forecast cone generated is not very good. That's the whole reason behind the Bayesian tear sheet, after all! But, the Bayesian tear sheet's forecast cone is, well... terrible. You can't see anything!

I searched for a solution but couldn't find one, so I looked through the source on github and made it work in research. I also found a bootstrap multistrike cone that isn't used in the tear sheets as a bonus.

Hopefully someone will pick up what I have revealed here and show how to customize the charts. There are additional charts and options as well.

1 response

backtest used to generate graph above