Back to Community
Simplified Moving Alpha Tearsheet

Attached is my latest tearsheet from an algo I've been building upon since my last contest entry in October. The main changes since then were to further reduce my exposures to beta, individual sectors and common risk factors(ie: Momentum, Size, Short Term Rev & Volatility). I did not add any additional factors. I kept the number of positions constant with approx. 1,025 rebalanced daily @ 10,000,000 initial capital. As it stands now it does meet all contest entry requirements and round_trips=True were set.

I thought it might be of interest to some in the Q community. I've learned SO MUCH from all of you top contributors out there from various combination techniques to mitigating common/risk factors. As always, I appreciate all/any feedback/comments.

Thanks

Loading notebook preview...
3 responses

Hi @Daniel,

Very impressive strategy in my book. Congrats also on your recent continuous success in the Q Contest! I'm officially jealous! :) Is it a variation of this strategy perhaps?

This strategy looks somewhat similar to some of mine, though in many ways I think yours might be better. I think it would be cool if there was a way we could compare our return streams, to see how correlated they might be. I don't think this is currently possible however, since we don't have access to the other's backtest return stream.

My only feedback is the same as I try to give myself: What precautions have you taken to minimize the likelihood (and severity) of overfitting?

I'm also curious if you'd be willing to share which datasets you used for this strategy, and if you still have any time-period available for OOS testing?

Hi @Joakim,

Thank you, so far it's been a great learning experience. The datasets I used for this strategy I didn't use anything fancy just a mix of free Fundamentals(not factset) and Psychsignal. As for OOS testing I've only used Oct-18 to Dec-18 since that was when I last made modifications. I will continue to use Jan 2019 onward as paper trading testing.

Thanks to you @ Daniel for posting this. Congratulations on an excellent-looking result. The numbers give us a great new level of target to aim for, especially that max Drawdown, VERY nice ;-))