Let's say your historical tests show that your strategy outperforms the market within a given range of error. Can we necessarily assume that the strategy will hold in the future? I guess there are two sides. The assumption that your strategy is unknown thus not priced in vs. the assumption that your strategy is already priced in. Do conventional strategies continue to outperform the market?
Alan,
Unfortunately, backtests do not predict the future. The purpose is to explore and understand the behavior of your algorithm, but it isn't a crystal ball that will tell you how the algorithm will do in the future. The biggest risk with backtesting is overfitting your algorithm to the historical data. With paper trading (walk-forward testing), you can go further and try your algorithm against new market data every day. That way, you have new out of sample data every market day. Still not a crystal ball, but it helps you avoid overfitting.
thanks,
fawce
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 is a very important question. One of the big dangers with providing simpler tools to investors to trade algorithmically is the myth of the backtest. I can construct any number of backtests you want that would look absolutely incredible if they were realized performance.
One of the most important parts of developing a sound trading strategy is understanding who or what you are taking advantage of and trying to fully understand what you are doing before you start running backtests. Are you a liquidity provider? Or a trend follower? will implementing your strategy inherently reduce its effectiveness or is it the sort of trading where the more people doing it the better?
Quantopian is a neat service, but there are a lot of people on here, including a few quantopian employees (unfortunately), that miss the entire point of a backtest. The "sell in may and go away" strategy is the perfect example of a terrible approach to quantitative investing.
It would be nice if Quantopian offered a way to step through historical timeframes so that you could produce out-of-sample backtests. John and Ray are spot on. Perhaps I am missing an important feature, but it seems Quantopian is limited to running backtests up until today. It'd be fantastic to offer a sort of "historically simulated" paper trading feature that would allow people to develop their algo using, say, the first half of available data, and then "paper trade" it over the second half. The in-sample backtesting is not worth much, as both John and Ray pointed out.
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.