VIX, VIX Front Month Futures Empirical relationship

Since volatility is mean reverting over time, VIX futures have picked up significant volume since inception. Valuations of the futures however are very different from the spot index (details at CBOE VIX Micro Site). So just how different are they? More importantly, can we exploit inefficiencies seen in VIX with the futures?

Here I present some statistics between the VIX index and front-month VM (VIX Futures Mini) for the past year.

Basic comparison between VIX index, front-month VM



We can see that the cash index is more volatile than the futures, more often, and they pretty much move in line with each other over time, but not from day to day. 


Moving 20Day Correlatio
Average correlation since Feb. 2011: 83%


Here we see that day to day correlation does break down from time to time, but longer term correlation tends to stay in the positive.  

Moving 20Day Volatility Ratio (VM / VIX)
Average Vol Ratio since Feb. 2011: 86%

From the graph we can see that this ratio isn't evenly distributed over time. This is consistent with the general volatility sentiment for the past year. 

What if your edge isn't available, or that risk exposure becomes too great with VIX futures? The simplest solution would be to trade the SPX index options, and make that Vega.