So, since index volatiliy is trading richly...

Quote from TskTsk:

You'd known had you read the thread before trying to educate me about your "transaction costs" and "4% jumps".

But yeah Ill go put the trade on. And I'll watch out for those nasty "transaction costs", those things can really wipe you out.
More...

i was being slightly sarcastic.

i still don't know the tenor you're discussing, and at this point i'm not sure if you do either.

I was really going for was the whipsaw effect and you constantly being underhedged during volatile markets, that swag put into real words for me.
 
I did you a favor since i wasn't being clear before.

Here is a graph of implied vs realized vol. For Implied vol i used VIX as a plug and for realized vol I used 20D realized.

Spot VIX is the weighted avg 30d forward implied vol. I choose to use 20D realized because 30D forward includes weekends and if you subtract the weekends out you're between 20-22 trading days in the VIX calc so to incorporate the same # trading days I used 20D realized(i think this is right? you can correct me if i'm wrong for choosing 20D realized)

You need to keep in mind that the VIX represents what people are expecting for the next 30 days while the realized vol is showing what already happened in the past 20 days.

If the graph shows negative numbers it means the market realized more vol in the previous 20d then the VIX is predicting for the next 30.

I only did it up to 10/2011 because I had the data on hand and didn't want to go grab more. you can see that implied vol is typically higher than realized, but there are some occasions where realized spikes above (the negative numbers on this chart) I hope this is somewhat helpful.
 

Attachments

Quote from Jgills:

I did you a favor since i wasn't being clear before.

Here is a graph of implied vs realized vol. For Implied vol i used VIX as a plug and for realized vol I used 20D realized.

Spot VIX is the weighted avg 30d forward implied vol. I choose to use 20D realized because 30D forward includes weekends and if you subtract the weekends out you're between 20-22 trading days in the VIX calc so to incorporate the same # trading days I used 20D realized(i think this is right? you can correct me if i'm wrong for choosing 20D realized)

You need to keep in mind that the VIX represents what people are expecting for the next 30 days while the realized vol is showing what already happened in the past 20 days.

If the graph shows negative numbers it means the market realized more vol in the previous 20d then the VIX is predicting for the next 30.

I only did it up to 10/2011 because I had the data on hand and didn't want to go grab more. you can see that implied vol is typically higher than realized, but there are some occasions where realized spikes above (the negative numbers on this chart) I hope this is somewhat helpful.
More...

The VIX isn't the right number. He can't sell vol at the vix level. If he were trading varswaps your analysis would be better, but it overstates the risk premium.
Secondly, like you said, the graph doesn't give Tsk any insight into how much of the risk premium he might be able to expect to realize.
VIX spot is 24 right now; 1M SPX is 22. 2 Vols is a lot.

Better analysis is to take the 30 day straddle and the 30 day realized vol going forward.
 
Quote from newwurldmn:

The VIX isn't the right number. He can't sell vol at the vix level. If he were trading varswaps your analysis would be better, but it overstates the risk premium.
Secondly, like you said, the graph doesn't give Tsk any insight into how much of the risk premium he might be able to expect to realize.
VIX spot is 24 right now; 1M SPX is 22. 2 Vols is a lot.

Better analysis is to take the 30 day straddle and the 30 day realized vol going forward.
More...

ok.

so for this we are going to assume both options are sold ATM or 100% moneyness, correct? i'll put this together shortly.

Also, you want to see the forward 30d vol at the same point in time as the 30d straddle so that you can see if the vol you sold is higher or lower than what was actually realized. Correct?
 
here is what we have been discussing. the implieds are 30d and the realized is 30d forward.

These graphs will help you see where you sold vol vs what vol actually realized.

any comments on changes?
 

Attachments

Quote from Jgills:

here is what we have been discussing. the implieds are 30d and the realized is 30d forward.

These graphs will help you see where you sold vol vs what vol actually realized.

any comments on changes?
More...

Yup. That's the graph. What does it mean?
 
Nice. I also found this graph lying around, from an old Goldman paper. The median spread is +2.7 percentage points from 89-05. It would be nice knowing the percentage spread in JGills data as well.

30D%20forward%20vol%20vs%20straddle.JPG
 
Quote from TskTsk:

Nice. I also found this graph lying around, from an old Goldman paper. The median spread is +2.7 percentage points from 89-05. It would be nice knowing the percentage spread in JGills data as well.

30D%20forward%20vol%20vs%20straddle.JPG
More...

Median isn't relevant because its the tails that kill you. The DAX paper you cited even mentioned the kurtosis of the return distribution.

But the conclusion is the same. The premium is generally there. When vol is high, the premium might become negative.
 
You must log in or register to reply here.
Back
Top