Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Dynamic update of factor parameters based on past portfolio value performance
Bernard Madopp
posted Aug 6, 2020

Hi,
I am trying to change parameters to compute a factor using past performance but I dont find a way to do it. For example I tried this

PV.append(context.portfolio_value)
NS=10
if len(PV) > NS:
if PV[-1]/PV[-NS] < 1: par1+=2

but the backtest results are not affected(they should depend on par1, but practically the change in par1 is not reflected in the backtest results).

This would be very important to obtain robust strategies, which are not ovefitting data. It is sometime called walkforward optimization, but in this version I am trying to implement would be just a dynamic adaptation.
It seems like zipline is not able to adjust factor parameters dynamically, and that factors are computed statically just at the beginnig of the backtest.

Thanks

2 responses
Bernard Madopp
Aug 9, 2020

To achieve what I mentioned is very important to obtain robuts oos performance and avoid overfitting. I am surprised there is no answer. Static factors are prone to overfitting and selection bias, so instead of wasting time testing many factors this functionality should be implemented.
It is would be a basic version of reinforcement learning

Dan Whitnable
Aug 9, 2020

Pipeline factors cannot be 'adjusted' based upon the current portfolio value primarily because pipelines are run asynchronously to the backtest. Typically, a pipeline will calculate 6 months of factors at a time. The backtest hasn't calculated the portfolio value yet.

One approach is to use pipeline to simply fetch data. Do all your calculations outside of pipeline. This isn't feasible for all types of calculations but may be in specific instances. Try it out. I would suggest putting all the calculations in the before_trading_start method. It get's more processing time than handle_data or a scheduled function. You can check context.portfolio.portfolio_value and store state as a context attribute.

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.

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.