SPX/VX

Discussion in 'General Discussion' started by DavidF, Nov 10, 2016.

  1. DavidF

    DavidF Well-Known Member

    A set-up I´ve backtested and traded a fair bit is short ES futures/long XIV, mostly in a ratio of 5:1. It performs paticularly well when SPX is flat and VIX is low with steep front curve on VX futures. I´ve also run it through major up and down moves (e.g last Aug 23rd) and it holds up well. On strong up moves it can fall behind but catches up once SPX stabilises and the front month futures continually converge to a low spot VIX.

    I did it through the election and with all the volatility the ratio is remarkably stable (calculated hypothetical XIV values based on VX futures during the overnight crash). The advantage of the set-up is max loss is very limited to the downside (e.g., $45k XIV hedged by $220,000 SPX so well protected to downside). I only put these on when differential between 1st and 2nd month VX future is above a threshold and/or before a binary event where I expect vol to collapse.

    Which brings me to my question:- based on historical & current volatility data (incl. VIX, VX, VX1data) wouldn´t it be possible to set-up software where you can model VX/SPX ratios in the same way as we already do with our option trading? The fundamentals are the same, you´re hoping that volatilty collapses quciker than the SPX moves but your shorting volatilty without the downside risk that comes with our set-ups.

    Any thoughts much appreciated
     
  2. ACS

    ACS Well-Known Member

    Was it back tested through 2007-2009?
     
  3. SVL

    SVL Well-Known Member

    David, Why XIV and not VX future ?
     
  4. SVL

    SVL Well-Known Member

    I used to trade volatility and followed a Vol Trader on seekingalpha.com – Itscalledcommonsense.
    I still read some of his comments and he uses a pair trade VX/ES and his performance is shown in the attached file.
    http://seekingalpha.com/user/1112012/activity-feed#view=comments_activities
    “I mean buy 2 VX front month and buy 1 ES. So that is $45k VX and $100k ES. You can do 1 of those combos or more. If you do 100 VX the proper ratio is 57 ES, as of right before the close.
    The hedge ratio between VX and ES changes as the days till expiry and the $ values of each change. So the appropriate ratio changes from time to time.
    If I give a specific trade idea that trade is only good when I give it and that specific trade may not be best even for a similar situation at a later date.
    Long basis trade with ES hedge are very high expectation at times like this. ES is there to reduce variance.”
    Itscalledcommonsense.png
     
    DavidF likes this.
  5. DavidF

    DavidF Well-Known Member

    ACS, regarding backtesting 2007-2009, in short no but I wouldn´t place it when there´s high vol/backwardation or low positive diffential between front month futures. However it works very well in fast down moves as XIV losses, although sudden & severe, should be more than offset by short SPX x 5 (e.g 10% drop in SPX equiv. to 50% fall in XIV).

    SVL, you could short VX but I prefer the limited downside of XIV and decreasing value with increasing volatility. Also find the bid-ask/volume on XIV ok. There are some good academic articles on VX set-ups. mostly calendar type trades with predefined rules. I´ve also done these types of trades but prefer the ES/XIV set-up
     
  6. DavidF

    DavidF Well-Known Member

    SVL, thanks a lot, exactly what I was looking for. Interesting that he/she is using 45/100 ratio and long VX, looks like Brexit was a challenge which probably indicates a high VX/SPX ratio wasn´t used.

    The ideal conditions for the trade I put on is extremely steep contango and low spot VIX (major VX loss) with SPX flat.

    I use a 1:5 ratio based on this:-

    http://seekingalpha.com/article/1932831-are-the-vxx-and-xiv-just-providing-leveraged-market-exposure

    However this ratio changes markeldy and it could be a lot better if there was a way to model the SPX/VX ratio based on current conditions. There are a lot more similarities than differences between this type of trade and an M3 IMO.
     
  7. SVL

    SVL Well-Known Member

    David, XIV is a byproduct of VX1/VX2 and that's why I thought it make sense to trade ES/VX1 as a pair as it virtually could traded 23 hours a day and entered/exited at the same time. There is still a risk with short volatility ETF/ETP that it could down to zero and will be liquidated in case of really severe market crash .
    I never traded that pair myself as I didn't have OV at that time but it might be a good field to explore and back-test for anyone who still has stomach for such wild swings in P&L.
    The other very good volatility trader is Stephen Aniston ( vixcontango.com). But he is more kind of TA based swing trader.
    Good Luck with volatility trading.
     
  8. DavidF

    DavidF Well-Known Member

    One other point SVL, the way they do it, long VX long ES is very different from short VX short ES and is more (a) a bet on a vol spike and or (b) best when futures are in backwardation. It could also work well in 2011 when there´s a lot of chop. I think it would be harder to time it unless there´s a bear market and XIV is losing both due to backwardation and volatilty drag. However returns look great so something I´m prob missing.
     
  9. DavidF

    DavidF Well-Known Member

    SVL. Agree, XIV does carry risk, but in a crash that severe the short SPX returns should crush the XIV losses. Correct, SVXYand XIV is rolling 30-day exposure to VX1 and 2, just meant that upside losses get worse with each % increase in volatility. VX1 hit 63 or so in 2008, could be very nasty loss if you were leveraged up.

    Thanks a lot for your input.
     
    Last edited: Nov 10, 2016

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice