Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Bug/Inconsistent Backtest Results

I am getting inconsistent backtest results when I include the following line. The backtest settings are identical... full backtest with minute data from 1/14/13-1/14/15.

record(leverage = context.account.leverage)  

I've included a simple example, first with the line commented out.

6 responses

Now here is the exact same code under the exact same testing conditions with the line uncommented.

interesting. To me it looks like the algo trades every minute with record enabled but only at open without it.

So any update on this? Should we assume minute backtest results to be inaccurate unless the record function is used in the algo?

I missed this post the first time, sorry about that. I just reproduced it and wrote a bug. I don't know yet what's causing it. I don't think it's record() itself that's triggering the problem. If I record something else the result doesn't change. The problem is somehow tied to context.account.leverage.

I'll ask a dev to look into it and come back here.

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.

This bug is now resolved in production.

It did not affect any real money algorithms. It did affect backtests and paper-traded algorithms in certain situations (like the example above).

If you're interested in the details, here we go: When an algorithm accessed context.account, it was incorrectly preventing the context.account values from updating correctly. As a result, the context.account values were generally being updated daily as opposed to minutely. This would generally manifest itself when using the order_percent() or order_target_percent() commands because they rely on an accurate and updated context.account object. It would also manifest itself when an algorithm was making decisions based on context.account.leverage.

Thanks for the update Dan.