Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
New to Quantopian. Need comment on simple XIV/TLT Strategy

Hi,

I'm new to Quantopian. I put up a simple XIV/TLT strategy and would need some comment.

The strategy comes from reading this article:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2369239&download=yes

Instead of using the future term structure, I think I'll just use the SPY MA as an indicator. The rules are simple:

Rebalance weekly.  
If SPY above 200 DMA:  
    long 40% XIV, 60% TLT  
else:  
    long 100% TLT  

The result is, to my surprise, very promising. DD is okish, at least to me.

I have some questions:

  1. I think my code is pretty naive. Can you please let me know if there're any obvious mistakes in it?
  2. I want to stress test this. I have a set of XIV simulated EOD data all the way back to 2000. I want to see how it behaves. I know how to use fetch_csv, but I do not know how I can put the simulated XIV price to the data.current[sid(XIV)] where the data does not exist after 12/1 2010. Can someone let me know how to do it?
  3. I found the performance deviate a lot simply by choosing different day of the week (or hours of the day) for rebalancing. Is there any way to reduce this uncertainty? Doing more rebalance helps but I do not want to over do it.
  4. I think simply doing 40%/60% should not be the right way of doing things. Can anyone let me know if there's a better way?

Thanks.

3 responses

Perhaps you may have a look at this existing thread:

https://www.quantopian.com/posts/for-robinhood-trading

Similar returns in comparison to a buy and hold XIV.

Tim,

Thank you very much.

Georges,

I guess the DD is the major concern.

Still wonder how I can use the simulated XIV data to backtest this thing...