Sorry for not linking the actual backtest, but I did not know how to share it while withdrawing the code. This is from January 1, 2011 to December 30, 2016 with a maximum drawdown of only 8% and constantly beating the benchmark of SPY. Let me know what you think and thanks for checking this out. After multiple versions this has been my most impressive to date.
How often does it rebalance a portfolio
What type of loss stopping procedure is impleted moving averages or % of portfolio or position
Hi Nathan,
A few things consider:
- What gross leverage are you using? I'd suggest testing at 1.0.
- Run a backtest as far back as the data will support (without tweaking your algo), and see how things look with pyfolio. This is one way to see if you've over-fit.
- Consider if you could get beta ~ 0. There's a risk that you over-fit to the rising market. As I understand, Quantopian is only interested in the pure alpha part of a strategy; there's no money in beta, from a hedge fund standpoint.
- To what extent did your workflow and resulting "product" conform to what is described on A Professional Quant Equity Workflow? Going forward, my sense is that a preference will be given to conforming quants and their algos.
- Are you using all equities? How many? ETFs, too?
- Scalability. My sense is that Q doesn't want to mess around with small allocations, so you might consider the capacity beyond $1M in capital.
- Generally, Q says "We look for robust high Sharpe ratio strategies that perform well out-of-sample" (see blog post). I think this means at least 6 months of paper trading, followed by a seed money allocation of ~$100K, which could then be ramped up gradually. My hunch is that to get to any scale, one would need several years of real-money trading, but it all depends on the risk they want to take. The point is that if there is any hint that you've over-fit, then your chances of getting an allocation will be diminished.
Hello Nathan,
Really, the most important thing will be how it performs out-of-sample. You really have to be careful about overfitting when you are doing backtesting. The out-of-sample testing will tell you a lot about whether or not your algo is overfit. When I read your comment, "After multiple versions this has been my most impressive to date," I worried that you might be falling into the overfitting trap.
Unfortunately, we can't really evaluate it for the fund with that image. It's not enough information. When we're looking at algos for the fund, our best tool is the tearsheet generated by pyfolio. Tearsheets don't include the code, so they are often easier to share. If you're engaged in a pair trade and you don't want to reveal the pair, you can run a version of the tearsheet without positions (though that makes the tearsheet a lot less informative about things like position concentration risk). Can you run a 10-year backtest? A tearsheet of a 10-year backtest is a great evaluation tool.
Disclaimer
The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.
Please sign in or
join Quantopian to
post a reply.
Need help?
Contact support.
Company
Learn & Support
Notices
Follow Us
The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian.
In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.