June 21st, 2005, 11:37 pm
quantstudent19> i cant really imagine a situation where IV would move up that much while realized volatility is zeroLet's assume for the sake of argument that the above is true. HV can move within the range of the upside & downside breakeven points without crossing outside. If such a position is held to expiration then it would expire worthless. However, if such movement within that range caused IV to increase more quickly than theta pushed the payoff curve towards the expiration shape (i.e. perhaps before the knee in the theta curve about 100 days out), then liquidating the position prior to expiration could result in gain.Next, as for the issue of IV depending closely and exclusively upon HV, I'd be grateful to be referred to literature which helps to reinforce this dependancy. I realize that B-S assumes they are closely related, but is not the relation rather imperfect in practice?For classes of underlyings in which options are primarily used to maintain some sort of hedge for the life of the position, perhaps it might be reasonable to assume that the relation between IV & HV tends to mean revert in a tighter manner? (just speculating)However, consider classes of underlyings in which options are often used for other than continuous hedging of the underlying:*) putting on a hedge as a substitute for liquidating an underlying position (i.e. costless collars), when such liquidation would be problematic (i.e. SEC reporting of insider transactions, or restricted/locked-up shares);*) as a means of quickly monetizing an iceberg order, with the options position undone gradually as the iceberg is worked;*) for creating synthetic futures;*) for speculative purchases of naked options by investors.The point of these examples is to ask whether it might be reasonable for supply/demand of options to move in a manner which occasionally drifts out of the usual correlation with the supply/demand of the underlying?For which categories of underlyings do you tend to monitor the correlation between IV & HV, where you observe that it tends to be rather tight? When I run the BB HIVG function on US equities indices, the correlation between IV & HV does not appear to be quite so strict. I do see some mean reversion there, but it does not seem to be quite so immediate.
All standard disclaimers apply, and then some.