Implied Volatility autocorrelation

Quote from cosine:

No, it's not enough to know when IV is relatively high, or relatively low.

If you expect clustering or mean-reversion in IV, so does the market. Options are priced consequently. You will have to beat the volatility term structure in order to make money, by anticipating implied vol more accurately than the term structure does.


Then it all comes down to ivol term structure modeling, which yes does usually require a Ph.D, unless you've learned how to solve PDEs in your finance undergrad at bschool.


Also, the 'direction' of the underlying doesnt have much to do with trading options.

And by the way, Forest Gump was not dumb.
More...


Agreed

Maw

Ps:Excepted for "direction of the underlying", cause every options are path-dependent.
 
Quote from cosine:

No, it's not enough to know when IV is relatively high, or relatively low.

If you expect clustering or mean-reversion in IV, so does the market. Options are priced consequently. You will have to beat the volatility term structure in order to make money, by anticipating implied vol more accurately than the term structure does.

Then it all comes down to ivol term structure modeling, which yes does usually require a Ph.D, unless you've learned how to solve PDEs in your finance undergrad at bschool.

Also, the 'direction' of the underlying doesnt have much to do with trading options.

And by the way, Forest Gump was not dumb.
More...


all my option trading involves a directional bet on an underlying.

you must mean direction does not have much to do with trading options relating to strictly volatility trading? even trading vix future spreads require a directional bias as one part of the reason to get into a switch. can you give a simple example of how one would trade an option for a positive p&l without a directional bias? of course i am speaking from a retail standpoint (thus no delta neutral trades).
 
Quote from rosy2:

download R. its free and will do everything.


>
> query "s=^VIX&a=11&b=1&c=2006&d=0&q=31&f=20008&z=^VIX&x=.csv"
>
>
More...

query = paste("s=^VIX&a=11&b=1&c=2006&d=0&q=31&f=20008&z=^VIX&x=.csv")

nitro
 
Quote from rosy2:

download R. its free and will do everything.


> library("fImport")
> query "s=^VIX&a=11&b=1&c=2006&d=0&q=31&f=20008&z=^VIX&x=.csv"
> VIX= yahooImport(query)
> pacf(VIX@data$Close, na.action = na.pass)
More...





Quote from nitro:

query = paste("s=^VIX&a=11&b=1&c=2006&d=0&q=31&f=20008&z=^VIX&x=.csv")

nitro
More...

I downloaded R and it doesn't seem as simple as suggested.

1st line:
"Error in library("fImport") : there is no package called 'fImport'
"

Anyone have any suggestions on how and where to get the library functions, fImport,
yahooimport(), etc for WINXP? And it would be great if you could be specific; not just point to a generic reference page.



TIA
 
Quote from dtrader98:

I downloaded R and it doesn't seem as simple as suggested.

1st line:
"Error in library("fImport") : there is no package called 'fImport'
"

Anyone have any suggestions on how and where to get the library functions, fImport,
yahooimport(), etc for WINXP? And it would be great if you could be specific; not just point to a generic reference page.



TIA
More...
You forgot to download the packages, otherwise known as contrib.

On the R program packages menu, select set CRAN mirror. Choose somewhere close to you.

Once that is done, select the packages menu and install packages. I just install them all (by selecting them all) so I don't have to bother with it again. Installing all of the packages on a high speed internet line will take a couple of hours.

Then try again.

nitro
 
Quote from nitro:

You forgot to download the packages, otherwise known as contrib.

On the R program packages menu, select set CRAN mirror. Choose somewhere close to you.

Once that is done, select the packages menu and install packages. I just install them all (by selecting them all) so I don't have to bother with it again. Installing all of the packages on a high speed internet line will take a couple of hours.

Then try again.

nitro
More...

You are awesome. Thanks. I wholeheartedly agree with RTFM, but I wish some of these programmers would be a little more documentation friendly (like maybe a 1 page visual tutorial on getting started and running, rather than pages and pages of verbose technobabble).

P.S. If it helps anyone else on their 1st attempt (WIN XP), I just simply typed
update.packages() at the prompt
(which popped up a list of packages/servers to select -- I selected all).
Took a while as Nitro mentioned, but script ran like a charm.
 
Quote from dmo:

If you want to trade IV, then you have to become intimately familiar with its behavior. The IV of each contract has a distinct personality. Choose one and target it exclusively for a while. Get to know its personality inside out, backwards and forwards.

Let's start with the vix, since it's calculated for you and so much data is available. The first thing to know about the vix is that every time the underlying ticks up, the vix ticks down. Every time the underlying ticks down, the vix ticks up. That is true 99% of the time. I'm attaching a one-minute chart (each bar represents one minute) comparing the E-mini (top) to the vix (bottom) over the last 4 days.

Surprising how rigid that relationship is, isn't it? But that's not true in all contracts. Each contract has a completely different and distinct personality and behavior. Chart it against the underlying, against anything else you can think of. Compare the ATM vol against the otm put vol and otm call vol under different scenarios.

Most of this you'll just have to do on your own. If you do it enough, you just may begin to see recurring patterns and correlations. If you do, this is gold, because unlike iron condors etc., you're now looking at something that not everybody in the world is looking at.

Look at the chart I posted. Does anything stand out? Notice that on Friday after 11:30 the e-mini rose steadily without a corresponding drop in the vix. That's at least a bit unusual. Does it mean anything? Can you play it?
More...



Thanks for this post because it clarifies what you meant in other treads when you said that one has to look for anomalies.
I have several questions:
1) The normal relationship is for the VIX to tick down when the e-mini ticks up. Therefore, the VIX and the e-mini are supposed to be negatively correlated. However, can one contract be thought of as the independant variable while the other be thought of as the dependant variable? Which would be which?
2) Could we think that the fact that VIX is mean reverting while the e-mini is not would explain why the normal relationship between the two sometimes gets out of line?
3) I don't understand why each contract would have a "completely different and distinct personality and behavior", could you rationalize this? Could you show us a graph showing another contract that has a different relationship with its implied volatility?
4) Knowing that on friday afternoon, vix did not trend down while the e-mini was going up, would you consider an e-mini option to be relatively expensive? then, how about shorting the vix as well as shorting the e-mini?
5) I notice that an a longer time frame, at the close of friday, the e-mini is higher than on the opening of thursday. We would think that the VIS should be lower at friday closing than it was on thursday opening, but it is not. Is what we described as the "normal relationship" doesn't hold on longer time frames?
6) Could you point me to somewhere I could get historical data to further analyse the relationship between the vix and the e-mini? Also, is there a possibility to have free tick data to follow the evolution of the e-mini in comparison to the vix?
7) How is the volatility skew supposed to evolve?
 
Quote from Rudolf13100:

Thanks for this post because it clarifies what you meant in other treads when you said that one has to look for anomalies.
I have several questions:
1) The normal relationship is for the VIX to tick down when the e-mini ticks up. Therefore, the VIX and the e-mini are supposed to be negatively correlated. However, can one contract be thought of as the independant variable while the other be thought of as the dependant variable? Which would be which?
2) Could we think that the fact that VIX is mean reverting while the e-mini is not would explain why the normal relationship between the two sometimes gets out of line?
3) I don't understand why each contract would have a "completely different and distinct personality and behavior", could you rationalize this? Could you show us a graph showing another contract that has a different relationship with its implied volatility?
4) Knowing that on friday afternoon, vix did not trend down while the e-mini was going up, would you consider an e-mini option to be relatively expensive? then, how about shorting the vix as well as shorting the e-mini?
5) I notice that an a longer time frame, at the close of friday, the e-mini is higher than on the opening of thursday. We would think that the VIS should be lower at friday closing than it was on thursday opening, but it is not. Is what we described as the "normal relationship" doesn't hold on longer time frames?
6) Could you point me to somewhere I could get historical data to further analyse the relationship between the vix and the e-mini? Also, is there a possibility to have free tick data to follow the evolution of the e-mini in comparison to the vix?
7) How is the volatility skew supposed to evolve?
More...

Rudolf - this is really the point at which you have to start observing this stuff for yourself and coming up with your own answers. Once you've studied a little and come up with your own observations, feel free to come back and ask what they mean.

VIX charts are widely available at msnmoney and other publicly-available internet sites. You can also get free tick data from realtick.com if I'm not mistaken. The first thing you need to do is set up your charting capabilities so you can take a detailed look at things.

Can I rationalize the different skew and IV behavior of the various option contracts? No, I'll leave that to the professors. I'm happy just to observe and learn what the normal patterns are so I'll know when they get out of line. That's good enough for our purposes. But that said, it's worth thinking about what most market participants are afraid of. In the stock market, the vast majority of participants are happy and relieved if the market goes up, and terrified of the market going down. That is reflected in the option skew, and in the inverse relationship between the VIX and its underlying. In crude, many more market participants are worried about the price going up, and are relieved when the price goes down. So you would not expect crude options to behave the same as S&P500 options.

As for your observation that "I notice that an a longer time frame, at the close of friday, the e-mini is higher than on the opening of thursday" - That's thinking and looking for yourself Rudolf - which is good! However, in this case I disagree with your observation. The apparent opening of the VIX around 20.2 looks to me like an opening data glitch, which is not uncommon in the VIX and in this case is related to the opening gap. I would ignore the first minute bar and look at the second bar, which shows the VIX opening at about 21.15, which is higher than the Friday close of 20.66.
 
Quote from dmo:

Rudolf - this is really the point at which you have to start observing this stuff for yourself and coming up with your own answers. Once you've studied a little and come up with your own observations, feel free to come back and ask what they mean.

VIX charts are widely available at msnmoney and other publicly-available internet sites. You can also get free tick data from realtick.com if I'm not mistaken. The first thing you need to do is set up your charting capabilities so you can take a detailed look at things.

Can I rationalize the different skew and IV behavior of the various option contracts? No, I'll leave that to the professors. I'm happy just to observe and learn what the normal patterns are so I'll know when they get out of line. That's good enough for our purposes. But that said, it's worth thinking about what most market participants are afraid of. In the stock market, the vast majority of participants are happy and relieved if the market goes up, and terrified of the market going down. That is reflected in the option skew, and in the inverse relationship between the VIX and its underlying. In crude, many more market participants are worried about the price going up, and are relieved when the price goes down. So you would not expect crude options to behave the same as S&P500 options.

As for your observation that "I notice that an a longer time frame, at the close of friday, the e-mini is higher than on the opening of thursday" - That's thinking and looking for yourself Rudolf - which is good! However, in this case I disagree with your observation. The apparent opening of the VIX around 20.2 looks to me like an opening data glitch, which is not uncommon in the VIX and in this case is related to the opening gap. I would ignore the first minute bar and look at the second bar, which shows the VIX opening at about 21.15, which is higher than the Friday close of 20.66.
More...

Ok, I will study this relationship and come back well I know a little more. Thanks for all your input.
 
You must log in or register to reply here.
Back
Top