Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Unaccounted for stock split <resolved> it was BRK-A

Something happened on the week of June 9 2014, unfortunately I can't actually say which stock it is because of the new transaction page having no sort function.

6 responses

June 9, 2014 stands out in the ledgers of stock splits as the day that AAPL split 7:1. Not sure thats the issue here but that date is suspicious.

I ran a code where I just go long AAPL, nothing happened.

Thanks for your patience as we improve the new backtest analysis page.

  • One option is to access the old transaction and positions table view. You can still access the old page by appending /old to the end of the backtest page URL.
  • Positions and transactions are also programmatically accessible via research using the positions and transactions attributes on the backtest object. You could search for the outlier in a notebook in the Advanced tab of the new analysis page.
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.

The problem is with BRK_A data on June 16, 2014. Some of the data is incorrect and reduced by a factor of 10,000. The algo stumbled upon that data and bought at the incorrect price of $18.932. It should have been $189320. That caused what looked like a 10,000x gain by the end of the day.

Good catch. That data probably has been wrong for awhile.

Attached is a notebook showing the bad data (algos and notebooks use the same data).

perfect

@Quant Trader I agree that the new backtest screen, while better in some ways, does lack in others. Being able to sort and filter the transactions is a big help when troubleshooting. One doesn't look at them a lot but knowing the individual trades is invaluable when debugging, or validating, algo behavior. I ended up scrolling through the (very long) list of positions to find one that had an abnormal gain. That was how I narrowed it to BRK_A. With the old backtest one could easily sort by gains in a few clicks.

@Josh Payne Thank you for posting the trick of appending /old to the backtest URL. That sort of allows the best of both. I understand the focus of the backtest output is to hone the trading strategy. However, the backtest is also often used to simply troubleshoot and debug one's algo.