Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Arbitrage Strategies of Crude Oil Futures Contracts

Hey guys, I am a newcomer in this quant Community-town.

In this year, on Mar 26th the Shanghai Inter Exchange Market would, the first time, list Crude oil futures in its trading board, code number:SC. I have to do some tests for my arbitrage strategies, and the best historical data I can get and the test platform must be those 'CL' futures contracts in CBOT.

Here, I would like to share an alternative method for chasing the price signal. The logic behind this method is very simple, to our target 'CL', I would like to compare the current price trend and its volume within 16-20 days together, with two particular historic periods which I have chosen since the very beginning.

For instance, I just picked up the prices moving curve [start='2011-02-05', end='2011-02-22'], where it's the starting point for the next uptrend also the stopping point for its previous temporary falling.

I'm gonna compare this period including its daily prices moving and volume with the future periods which I mean from 2013 to 2018. If there is a specific future period showing up a perfect or quite similar tendency in a particular ratio between prices changes and volumes changes (let's say 9:1). I would assume the "Yesterday Once More", and in a quite similar environment persons would almost do the same choices if the external signals are also similar. So far, It is very easy to understand I guess.

In addition, I also chose one more period from 2012 for my short selling plan. Once again, if the current period matches up with my chosen past periods, I would order to open a short or long position of CL contracts, and after a week I would do a reversed operation to see whether I can benefit from it or not?

Jan 01 2013 - Feb 27 2018, Total returns 74.09%, Sharpe 0.77. I would say it's not bad, at least I merely did the daily trading not the intraday trading, of course with a quite higher max down(22.44%) it's a problem.

You can say this methodology is insane and not logical, but in my mind the persons would always do the same things if they were in a same environment with those same signals. Yesterday, you passed a river and the weather was good, you would tend to take time to stop and admire the view, and the other you in the other universe when he passed a river and the weather was good, he also would tend to stop and Idk, smoke? It must be something very similar because the external signals were the same.

Ok, I think that's enough for my first post in this Q-comminity, so if u guys have any ideas about it, plz comments below, thx.

Sirui

1 response

Backtest: Jan 01 2013 - Feb 27 2018, Total returns 74.09%, Sharpe 0.77, Max down -22.44%.