Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
The Logical Invest Enhanced Bond Rotation Strategy
Chirag Agrawal
posted Apr 13, 2015

I took a bond rotation strategy from Ilya Kipnis at QuantStrat TradeR. He coded the strategy in R, and I simply translated it into Python. I've since played around with the strategy a lot by tweaking various parameters. What's attached here is the original strategy. Unfortunately, I don't think it's a good strategy for a few reasons. Here are two main ones:

  • If you change the reallocation scheduled function from month start to month end, the strategy does significantly worse (returns from ~100% to ~60% in the same time period.
  • If you change / add bond ETFs, the strategy again does much worse. This suggests that the original creator (not Ilya) overfit to securities that work well.

Any thoughts?

4 responses
Bharath Rao
Apr 13, 2015

If the equity curve is too good to be true, it's probably an overfit.

Kiran
Feb 27, 2016

The original strategy relies on the 4 assets (bonds in this case) being relatively uncorrelated. Which bonds do you replace with, and do they have the same correlations and volatility as the original ones?

Kiran
Feb 27, 2016

It's very sensitive to the Lookback parameter - changing it to either 60 or 120 degrades performance, suggesting curve-fitting.
- Are there utilities in Quantopian to test robustness of parameters? For example, sensitivity analysis of performance to the parameters ..
- Also, not sure if Quantopian has Optimization features (exhaustive or genetic)..

thx
Kiran

Grant Kiehne
Feb 27, 2016

Are there utilities in Quantopian to test robustness of parameters? For example, sensitivity analysis of performance to the parameters...

Kinda, sorta. You can vary the parameters yourself manually and then pull the backtest results into the research platform, using get_backtest(). Note that you can run as many backtests in parallel as you want, so this might be a viable approach. There is the so-called pyfolio tool, but I don't think it'll handle multiple backtests and perform sensitivity analysis across parameters.

Also, not sure if Quantopian has Optimization features (exhaustive or genetic)...

Not anything built-in. However, if you port your backtest over to the research platform, then I think you could try such a thing (although I fear it would get mighty bogged down, since you'll only be able to run one backtest at a time).

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.