Hello Everybody,
This is an example of one of the metrics used for algorithm performance assessment in the Quantopian Managers Program. It calculates the r-squared error of a regression through the portfolios equity curve. This measures how consistently an algorithm generates its profits over time. It penalizes large drawdowns, especially ones that take a considerable amount of time to recover.
This example calculates the cumulative returns stability at the close of every day. The current screening process only considers the final (most recent) value. You are encouraged to plug this into your algo to get a better idea of how it performs, a stability of returns > 0.5 is a baseline for fund consideration.
Best of luck in the Quantopian Open!!
David
