Quote from hlpsg:
This seems to be more true of the longer term options than of the nearer term. Why is this?
I've attached an example of the SPX volatility curves of different expiration months.
Even as far out as the MAR09, there is still some evidence of a smile on the upside. This smile is very obvious for months closer than MAR09.
I did the same for RUT, and the smile is evident in the SEP08 and OCT08 options.
For NDX, up to and including the OCT08 expiration, you can see evidence of a smile.
Quote from JohnGreen:
dmo--
I think the reasons for the skew are obvious. The market has dropped 10% or more in less than a week quite a few times, but rises more slowly than it falls. Most people are long stock and worried about crashes so they buy protective puts or put spreads. I trade options on the NDX and part of my position usually includes a few DOTM put spreads to insure against this very disaster. Like Atticus and Charles Cottle, I want to be ready for the steep drop (by being long the wings) whenever it occurs because it will rain money when that happens.
The cost of these puts is simply a relatively cheap insurance policy which totally hedges my portfolio against a 1987 drop.
Perhaps a more pertinent question would be "is the skew justified ?" In other words do stock index distributions exhibit a kurtosis that the implied skew suggests. From the studies I've done the answer is no, they don't. In fact if you analyse 25 years of data and strip out Oct '87 the Kurtosis is very close to a normal distribution (kurt 3).Quote from dmo:
If you ask me "why does the skew exist? Why are people willing to pay more for OTM puts than for OTM calls?" My first and most honest and most visceral answer would be "The desire to avoid pain." That's consistent with the concept of portfolio insurance because again, that whole phenomenon stems from the fact that the world is long stock.
Quote from Profitaker:
Perhaps a more pertinent question would be "is the skew justified ?" In other words do stock index distributions exhibit a kurtosis that the implied skew suggests. From the studies I've done the answer is no, they don't. In fact if you analyse 25 years of data and strip out Oct '87 the Kurtosis is very close to a normal distribution (kurt 3).
So the Put buyer is still paying a hefty premium for an event that happened 21 years ago. Justified ?