Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Live Trading: counter-trades
Umed Saidov
posted Apr 10, 2016
Seeking Help Tools and Tips

I was wondering how do you protect your algorithm from being taken advantage of during live trading sessions. A while back I watched a TED talk about how one set of companies "sniff out" trading patterns/signals out of the trading data and trade against them.

I thought I could use some sort of random function to introduce some variability in my purchase/sale patterns but that seem to be restricted.

The trouble with back-testing is that it does not take into account the actions of other players who are "listening" for trading patters in the live trading stream.... To the extent that this is true, your algos may not work in live trading despite producing good back-tested results.

Has anyone looked into this?

3 responses
Grant Kiehne
Apr 10, 2016

Regarding the use of a random function, it is not actually restricted, as I understand. Rather, you need to use a fixed seed for the pseudo-random number generator, so that backtests are repeatable. You can use this:

http://docs.scipy.org/doc/numpy/reference/routines.random.html

This may be blocked, however (probably unnecessarily):

https://docs.python.org/2/library/random.html

Umed Saidov
Apr 10, 2016

Thanks Grant.

I can see how a fixed seed could be useful for replicating results, but for live trading, could we incorporate a truly random number (system clock?) into the timing of trades?

My investment thesis is that as long as stock attributes are in the acceptable range, timing of the trade should not matter (they would not change from one minute to another anyway). In addition to eliminating the effect of timing on returns, randomization could protect you from others "reverse engineering" your trading algorithm. My concern is writing algorithms with predictable trading patterns that are easy to front-run or trade against.

Maybe the problem isn't as bad as I make it sound...

Grant Kiehne
Apr 10, 2016

I think that the fixed seed is required so that results can be replicated by Quantopian if your algorithm is live trading, I'm guessing for quality control (e.g. if a change is made to they platform, they can run algos before and after the change, to make sure that they aren't affected). I think that the terms of service may also allow them to run any algo on the platform and examine its "exhaust" (not its code) as a candidate for the hedge fund. So, they'd want results to be repeatable.

About Quantopian
Careers
Community
Events
Documentation
Academia
Lectures
Status
Facebook
Twitter
LinkedIn
YouTube
Terms of Use
Privacy Policy

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.