Want some nice stats for your investment strategies, but don't want to mess with the gigantic pyfolio package?
You're in luck!
This framework borrows from pyfolio but is drastically minimized to avoid bloat. No dependencies except for numpy and pandas (and matplotlib if you want the plots).
There's probably some more statistics you might want - please open an issue and tell me how ignorant I am of the One Metric to Rule Them All and I will likely listen to you and learn something - these are just the ones I use.
-
Prepare your strategy for evaluation by creating a
pd.Seriesof the daily returns of your strategy, and the daily returns of a benchmark strategy. -
Feed the returns to
create_tear_sheetfunction. -
Voila!
I've included sample strategy data (buy-n-hold AAPL as your strategy, buy-n-hold SPY as your benchmark) and a notebook for you to play with and see how it all works.
Have fun! And if you're just getting started with algotrading:
First, ask yourself if the allure of skimming a few pennies off the top of a corrupt system that will overtake your life, blacken your soul, and bankrupt you in the process is the best use of your talents. No? Good choice; back to Hacker News. Yes? Get a second opinion from your mother. I’ll wait.
~ "Doctor J"