VIX calculation

Archidamus

Member
Anyone here familiar with the Cboe VIX calculation? I have a question about it (http://www.cboe.com/micro/vix/vixwhite.pdf).

\[ \sigma^2=\frac2T\sum_i\frac{\Delta K_i}{K_i^2}e^{RT}\;-\frac1T\left|\frac F{K_0}-1\right|^2\; \]

When there are two consecutive bid's of 0 then the remaining strikes are not included in the calculation. Therefor all of the strikes below 1150 and above 3100 (at moment of writing this) are not used. The original formula from which the Cboe's was derived used an integral of prices from 0 to infinity, suggesting that all terms should be used. My question is, does this not give way to manipulation by adding/subtracting bids on the far OTM strikes to attempt to get them included/excluded from the calculation? Or am I misunderstanding the VIX calculation formula, and there is some weighting aspect that minimizes potential gaming?
 

StephenKim

Member
C++ Student
Is the VIX vulnerable to manipulation? Some think so: http://www.jgriffin.info/wp-content/uploads/2017/12/vix_pub.pdf

And of course there have been many examples of very widely known fixing/rigging scandals (LIBOR, FX), as well as numerous lesser-known incidents of manipulation, like: "banging the close" of a futures contract to influence its settlement price in order to affect the corresponding "TAS" (Trading at Settlement") futures contract; quote-stuffing; etc.
 

Archidamus

Member
Yeah, that paper has some interesting data and ideas. Looking back to the leveraged volatility product debacle in February, I would say its more likely than not that there was manipulation and 'banging at the close'.
 
Top