Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
questions on Machine Learning

HI, why is it that I find machine learning alpha combination models perform poorly in comparison to the simple addition after normalization approach? This appears to be true for all combinations of factors I've used. I'm using the same code presented in Machine Learning on Quantopian notebooks, by the way...
I'm using both the tearsheet and backtest results for comparison.
Thanks in advance!