You do realize that in case of options on futures (e.g. VIX futures in this case) there is no difference in pricing of American and European options?
I don't think that was the issue. I think it was that the volatility started to increase so the price on the front month options went up, whereas the projection was for decreased volatility in the back month and the price didn't go up.
Until QE2 is exhausted in a few months, there's no reason to think the market will go down. VIX could see 10 by then. Don't Fight the Fed.
" -------------------------------------------------------------------------------- Quote from chrismontez: I started trading options on the VIX when they first came out. I noticed that the near month calls were selling for more than the far month calls. I thought I had found a sure thing with a pricing error and bought spreads selling the near month and buying the far month, assuming at expiration of the near month calls the far month calls would be worth more because of the time premium left. Well it didn't work out that way at first with the near month calls going up in price and the far month dropping. I got cold feet and closed the trade when I got back to even. -------------------------------------------------------------------------------- Because VIX options are European style!" You can see this phenomenon very clearly in most European style futures options products--but specific nuances in the VIX options will drarf this difference. VIX options expire into cash based on the VIX index cash number calculated from a strip of a basket of SPX options from the two nearest term SPX expiries. Therefore, each VIX options expiry forward price is derived form an indexed implied vol calculated 30 days further down the SPX vol surface. In short, each VIX expiry has its own forward price. Calls may go up in one VIX month, and down in another, due strictly to delta, if the SPX implied vol relationships between expiries changes. This is not to be confused with the implied vols of the VIX options fluctuating, which also happens. "It's because different VIX futures are related between each other by root-time. So, UX1 vs UX2 "beta" will be around 1 + sqrt(1/12) ~= 1.3" What? SLE, could you explain your logic? If this is some method that you use to correlate SPX implied vols between expirations, then I feel your model is off. If this isn't a method to model the implied vol changes between the different SPX expiries, then your logic is flawed. Am I missing something? The VIX futures are calculated from the implied vols of their correlated spots SPX options vol surface.
jerkstore you are correct. for the futures most traders will find little logic in how they trade due to so many factors. also trying to game fairval is not recommended as mm's catch up within seconds. these are mainly used by market makers offsetting risk for vix options and the such. the balance of is institutions buying or selling in systematic fashion. very little edge here unless you want to dedicate 100% of your time. http://www.cftc.gov/dea/futures/deacboelf.htm note the amount of traders.
Why do you feel that this is "off'? In general, first principal component of the vol surface dynamics is a root time relationship. Implied "fair" variance (which is what VIX is) for each expiry follows that relationship very well. If you do any sort of term structure trades, you would normally put them root time vega flat (flat notional) for the same reason. A simple regression is all you need to check that and this root time relationship is true for almost every asset (excluding rates and commodities).