Trading pairs with options

Discussion in 'Options' started by oldnemesis, Feb 22, 2015.

  1. http://en.wikipedia.org/wiki/Pairs_trade

    The ETF's DBA and MOO are a natural pair since they track the price of agricultural commodities in the case of DBA , and the stocks of agri companies in the case of MOO.

    http://finance.yahoo.com/q/pr?s=DBA+Profile
    http://finance.yahoo.com/q/pr?s=MOO&ql=1

    For the most part these two track each other but have now diverged:

    http://finance.yahoo.com/echarts?s=...sons":{"MOO":{"color":"#cc0000","weight":1}}}

    I would like to create a position based on calls/puts instead of being short/long stocks but the question is do I create a position based on equal dollars in the two elements of the pair (as is done when going long/short stocks) or on the net delta of the two call/put positions??
     
  2. xandman

    xandman

    To get equal dollar positions you calculate the "Dollar Delta" or $Delta of the position.

    But pairs trading goes beyond simply matching dollar amounts (See Kaufman: Alpha Trading) and you want to factor in volatility which may or may not be correlated as well.

    You may want to look up "Dispersion Trading" which is the full application of these concepts. Otherwise, back of the napkin $Delta with puts and calls is one way.

    I'm am not even looking at those 2 ETFs but will tell you that they are too tightly correlated for anything more than intraday trading (probably HFT). So options is out of the question. You have to find really loose, but reliable correlations.

    Additionally, you are better off with the leverage of futures which pairs (essentially spread) trading really demands. I think I just blew up my IRA fading Joe Ross' newsletter on the (seasonal as opposed to statistical) Hog/Cattle spread. It really isn't as safe as people make it out to be. Deviations from a range or correlation breakouts ( think Bollinger breakout) usually means a run away futures market that could be locked limit for days in one, albeit wrong direction.

    It's a mini-black swan when an known/obvious correlation breaks.
     
    Last edited: Feb 22, 2015
    Baron likes this.