Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Task 4(b)- Growth Grade against RSI Traditional (14 Days)

I have plotted heatmaps of Growth Grades against the quantiles of RSI traditional using the mean returns. Growth grade is an asset classification and tells how the company is performing relative to its sector.

For the overall quantile analysis, the only robust results can be obtained for growth grades C, B & D as they have the highest number of observations, and of them all, C has the most number of observations in it. With this in mind, the First_30 mins, and overnight returns exhibit arbitrage opportunities. Specifically, the overnight returns for the lowest quantile stocks in growth grade C & D are quite impressive and tend to show some statistical relationship that could be exploited as a systematic strategy.

Moving on further with the analysis on sectoral level, there seems to be an obvious correlation between RSI and growth grade in all four types of returns. Even after accounting for the number of observations, one can hypothesize the existence of arbitrage opportunities.