How Do Options Make Predictions?

[stevegee58]

Quote from dmo:

Looking at the SPY options, I see that the 95 strike is trading at 31% IV, while the 60 strike is trading at 70% IV.

So if IV is the market's predicted volatility of SPY, then what does the market believe? That SPY's actual volatility will be 31%, or that it will be 70%?

You're right. I meant ATM IV.

This is one reason why it's so hard to make money with earnings straddles. Front month ATM IV is pumped up in anticipation of high actual volatility. I've watched these things and seen the actual move be very close to the "predicted" move.

Of course once the news is out, the IV returns to normal.

 

[uptik2000]

Forget all that research and science crap! Do what 95% of the people do

 

[Random.Capital]

Quote from stevegee58:

That's simple. Option premium is set according to IV. IV is the market's *belief* of how volatile a stock is gonna be within some time period.

I think it's a bit more complicated than that. The IV calculations are based on a whole mess of assumptions regarding the market makers' ability to hedge away unwanted risk.

IMO it would be more accurate to describe it along the lines of "IV is the market makers collective belief about how much short-term volatility can be adequately dealt with before the risks become untenable".

 

[dmo]

Quote from black diamond:

Why do I get the annoying feeling this is a quiz question and not a real question? I've seen some of your posts before, I think you know this! But I'll answer you anyway to avoid having to read through pages of silly guesses to get to the good stuff.

Vol skew comes from using a model that assumes normally distributed returns on an asset with a non-normal return distribution. If the return dist was normal then you could have the same implied vol at every strike . But since stock returns have a greater probability of a large move down than a symetrically large move up, low strikes need to have higher implied vols than high strikes. This is the impact of skewed returns (smirk), there is also a distortion from fat tails(smile).

Sorry for being so unclear. My question was meant neither as a real question nor a quiz question but rather as a rhetorical question. Its purpose was to illustrate some of the problems with the common belief that IV is the market's expectation of future volatility.

You probably know that the model assumes lognormal returns rather than normal returns. I agree that even a lognormal probability distribution does not reflect reality. But I believe the skew is much more a reflection of the greater demand for downside insurance (OTM puts) than for upside insurance (OTM calls), and really has little to do with probability distribution.

 

[black diamond]

Quote from pinthestrike:

Not to be an ass, but these "violations" only occur in listed markets. Why is that? It's due to regulatory or investor intervention. Either the regulator puts a restriction on the short or the client takes possession of the security on loan. Voting rights? Comical.

Persistent, anything beyond one trading day in duration.

I must admit, this place is far more lively than Wilmott.

That is a perfectly logical explanation.

Another logical explanation is that there is a limit to the capital arb traders will throw in to maintain parity, and sometimes informed traders temporarily overwhelm it. This is more likely in exchange traded products because of the anonymity there. If a dealer in an OTC market notices he is on the wrong side most of the time he trades with you I would guess you are going to start getting shitty quotes after a while.

The research I mentioned and stock price moves after PCP violations seems to favor the second explanation over the first. But maybe you are right that persistent violations are something else - a day is a long time.

The voting rights idea is not as crazy as it sounds but I just doubt it actually happens often. There really are some differences in the voting rights from a stock and the synthetic. And there are times when the vote rights are really super valuable - think about the Bear Stearns vote on whether to accept the JPM terms. If the shareholders had voted to play chicken the bondholders would have risked losing probably tens of billions. But I would be surprised if this explains more than 1 in a million PCP violations.

Yes, I think people watching is better here than at Willmott. And if you want to be an ass by ET standards you have to try much harder.

 

[black diamond]

Quote from dmo:

Sorry for being so unclear. My question was meant neither as a real question nor a quiz question but rather as a rhetorical question. Its purpose was to illustrate some of the problems with the common belief that IV is the market's expectation of future volatility.

You probably know that the model assumes lognormal returns rather than normal returns. I agree that even a lognormal probability distribution does not reflect reality. But I believe the skew is much more a reflection of the greater demand for downside insurance (OTM puts) than for upside insurance (OTM calls), and really has little to do with probability distribution.

No problem. I have been working a lot and sleeping very little lately so I am probably too fried to get subtlety right now.

If you mean Black-Scholes I thought it was lognormal prices and normally distributed continous returns. Even if I am wrong my point is still that a single implied vol can't work at all strikes if the model distribution does not match the real distribution. Or the market's perception of the real distribution if you like that better.

But I think vol skew does have to do with the return distribution. There is greater demand for downside insurance because there is a greater probability of a downside move. If the world believed in a symetrical return distribution we would not have the asymetrical demand for insurance.

 

[shortie]

OP asks an innocent question and MAESTRO spills the beans about his super secret option strategy.

hilarious

but OP never gets his question answered as far as I could tell

 

[ML_QUANT]

Quote from MAESTRO:

May be this will help a bit

http://www.sciencedaily.com:80/releases/2008/03/080331110052.htm

A very interesting study indeed.
Meanwhile, do you agree that 100% of all physical nature can indeed have a mathematical solution(even if not known or discovered by human yet)?
On the other hand does the human mind fit within that physical nature with a 100 B. neurons that each have 2500 branches if you will?
I believe not, the human mind is not part of the physical nature hence the randomness and the inability to mathematically define it.
However there is indeed many patterns to their collective behavior contrary to their singular and solo actions and reactions which explains the school of fish or the flock of birds mindless following of the herd... I kicked off the same concept on the TS forum a couple of 2 years ago and no one caught up to it then.
The advance of TA is impressive and makes the scientists predictions of kids in the near future making Millions of $ daily at their home quite believable.

Back to the topic at hand, during the 86 era, when the INX was the only electronically instrument traded(I think), the attraction of the options was that they over reacted to any event of the underlying price and presented huge scalping opportunities which seems to have dissipated by now for obvious reasons and thus my silly question, where is the edge of trading options nowadays Vs. the underlying which is cheaper and faster to trade?

 

[ChocolateGirl]

Quote from shortie:

...but OP never gets his question answered as far as I could tell

That was my feeling too. After 100 responses, beside that some people can make money using the predictive power of options, what have we learnt?

Can someone summarize it? Thanks...

 

[dmo]

Quote from black diamond:



If you mean Black-Scholes I thought it was lognormal prices and normally distributed continous returns. Even if I am wrong my point is still that a single implied vol can't work at all strikes if the model distribution does not match the real distribution. Or the market's perception of the real distribution if you like that better.

But I think vol skew does have to do with the return distribution. There is greater demand for downside insurance because there is a greater probability of a downside move. If the world believed in a symetrical return distribution we would not have the asymetrical demand for insurance.

Right, lognormal prices. That's what I meant.

The skew in stock market index options doesn't change much, but the skew in other options - such as options on crude - does change a lot. Sometimes the OTM calls trade way higher than the ATMs while the OTM puts trade at the same IV as the ATMs. Sometimes it's the puts that are expensive and the calls that are cheap.

According to your theory, more expensive OTM puts would mean a higher probability of a move down. My observation is that the opposite is true. More expensive puts mean excessive bearish sentiment, which is bullish.