Estimating vol skew from historical underlying price data

[TskTsk]

Is this even possible? If so how does one go about doing it?

 

[sonoma]

Quote from TskTsk:

Is this even possible? If so how does one go about doing it?

No.

 

[spindr0]

Skew isn't a function of UL price

 

[TskTsk]

Gah, but I want to price OTM/ITM options based on UL history And I assume skew data is needed for that...

 

[newwurldmn]

You can estimate realized skew, but not implied skew. But for a retail account, skew is a tough asset to trade. You need a lot of leverage to make the vol difference worth your while.

 

[sle]

The answer is both "yes" and "no". There is a whole bunch of statistical studies you could do, depending on what you are trying to achieve.

 

[TskTsk]

Quote from sle:

The answer is both "yes" and "no". There is a whole bunch of statistical studies you could do, depending on what you are trying to achieve.

Care to expand a bit on this? My main objective is to price ITM/OTM options by looking at historical underlying prices, in order to do some backtesting. By the way after googling around a bit, I saw a post by you where you said you had an excel spreadsheet that does something along the lines of what I'm looking for. (I could have misread). Anyways, do you still have this? Would you be willing to share?

Quote from newwurldmn:

You can estimate realized skew, but not implied skew. But for a retail account, skew is a tough asset to trade. You need a lot of leverage to make the vol difference worth your while.

That's true. I'm not really looking to trade the skew, rather to backtest something which requires me having prices for OTM/ITM options.

 

[babutime]

Wow, now this is an excellent thread! I'd love to know that too. Damnit, too much partying last couple days, gotta get serious now.

 

[opt789]

You want to do back testing with regard to option price, and you are spending all this time trying to guess what the skew was based on the underlying?
Why not just use the actual historical option prices?

 

[TskTsk]

Quote from babutime:

Wow, now this is an excellent thread! I'd love to know that too. Damnit, too much partying last couple days, gotta get serious now.

Yeah, I actually found some papers on it. Here's one of them: http://www.math.nyu.edu/phd_student...="dupire skew from historical data bloomberg"

On page 24 its decribed how you can find the breakeven vol surface based on purely historical data. The assumption made is that deltahedging is a breakeven game, thus the term breakeven vol surface. Not sure if I understand the math behind it, so if someone who gets it could explain that would be nice. (It's on page 24)

Quote from opt789:

You want to do back testing with regard to option price, and you are spending all this time trying to guess what the skew was based on the underlying?
Why not just use the actual historical option prices?

I dont have the actual historical option prices.