Book And Research Paper Forum • Re: End-of-year Competition: David Orrell's q-variance paper in Wilmott

To answer the binning question, the q-variance equation doesn't say that v = sigma^2 + z^2/2, it says the expected value of v follows that form. The way it is derived is by looking at cases where z is near some value, say 0. So we are testing what it says on the can. One way to look at it is to e.g. set T=5 days and look at all the periods where z is near-zero. Then paste them together so you have a series of returns, which represent the system in this state. The variance v of the whole series i