I guess I would be comparing spread vol to price vol. In terms of my argument, I would assume that a company's debt and hence credit would be less volatility compared to equity. If I'm not mistaken, bonds do tend to be significantly less volatile than stocks.

Why is the implied volatility on the CDX index typically higher than the VIX? I would have assumed that moving up the capital structure would lead to lower volatility. Thanks.

<t>I am using OptionMetrics data for SPX options. When I use the implied volatility from the database to calculate the fair prices of options, I consistently get high prices (somewhat above the lowest asking price for both puts and calls). I am trying to figure out what might be causing these high p...

How are dividends accounted for (if at all) in the CBOE SPX index? In assessing volatility, skewness, and kurtosis, is it important to analyze the index's cumulative return (i.e. including dividends paid) or is merely the "price return" sufficient?Thanks.

<t>Underlying security is SPX, so that should work fine. I assume that I can fit this with a minimization of squares, correct? How is it ensured that the resulting distribution is a pdf? Does it even matter in terms of calculating skew and kurtosis if it is not exactly a pdf, i.e. does not integrate...

Anyone know how to ensure that the polynomial meets the definition of a pdf? Or, is it ok if it does not exactly meet all the pdf criteria? Can we still estimate the moments regardless?

<t>I think what was really confusing me was that I misinterpreted the fact that the prices must be the same that the expected return on the options must also be the same. Now I realize that this certainly need not be the case. I revisited Black and Scholes original paper and their CAPM-based derivat...

Oh ok. Gotcha. Thanks for clearing this up for me.

It just seems bizarre that one would pay the same amount for the two call options. I know this situation is highly unlikely, but am simply thinking about it as a theoretical exercise. As an investor, I would think the option on stock A is much more valuable.

<t>I have a quick question regarding risk-neutrality. In the Black-Scholes framework, the expected return on the stock is irrelevant to the price of the option. I'm having trouble wrapping my head around the following scenario: imagine call options expiring in several days on two identical stocks, e...

<t>Regarding the textread function, is there a way to import the data into a single matrix rather than separating it into three arrays? How come Matlab fails to import a ASCII text file with the date in the following format (19940105)? Is there a way I can just treat the date as a numeric data type?...

I think I understand everything now. I, however, have one last question. How do you go about ensuring that the fitted interpolated and then extrapolated polynomial is a pdf (i.e. integrates to 1, always positive, etc.). Thank you again for all of your help.

<t>Ok, thanks for the advice. I understand my kurtosis estimates will likely be quite poor; however, I am looking more for relative differences between certain periods rather than a precise number. Now, with your method, is the distribution you generate that of the underlying asset price. I'm wonder...

<t>Thanks. Interesting paper. Regarding my first post, I think I might have figured out my own question. When you calibrate a mixture of two lognormals to the current set of option prices, you get five parameters: the mean and dispersion parameters (alpha and beta) for each lognormal plus a weightin...

<t>I'm currently working on generating a series of option-implied skewness and kurtosis estimates using the Gram-Charlier method of Corrado and Su (1996). I have also looked at using a mixture of two lognormal densities to back out the implied probability density of the terminal asset price. My ques...