Can the following be thought of as "Dispersion Trading"? I sell Cash Secured Puts (american style) on a stock and collect the premium. I use that premium to buy DOTM puts options (european style) on the index the above stock is part of with a beta of 0,70. Both options have same expiration. Thanks
This is long correlation/ short dispersion. If you want to be long dispersion you do the opposite. Long the street vol and short the index vol. You should vega weight them if you are doing dispersion/correlation trading. IMHO.
The idea behind Dispersion is to buy vol in the components of an index for a lower cost then you can sell the component symbol vol (or the reverse). You are looking to do the reverse with one symbol with no process to lock in a profit. It does not mean you can't make money, but OTM index puts are skewed very high vs historical vol. The math does not support your expectations. https://www.math.nyu.edu/faculty/avellane/Lecture10Quant.pdf
My comment was not on the validity of the strategy. Just that I would not call it Dispersion. Dispersion requires a basket of buy/selling options vs a correlated index. E.G. buying vol in some bank stocks vs selling vol in the bank EFT because the basket is trading lower.
These days, dispersion describes a lot of trades, including a book where you selectively buy single name vol against the index. So the trade OP described can be classified as reverse dispersion.
No, you don't. It's zero-sum math. When you sell a cash-secured put, your required margin capital will grow by [(Stock*100) - Premium]. Your premium has already been used.
I guess we can call any strategy whatever we want, but without an instance of "dispersing", which requires a range of values, it make no sense. Buying vol in IBM and selling vol in the QQQ is a hedge but not a dispersion strategy. It would require a number of correlated symbols. The number required would be enough to create a statistical correlation.
Well, as I said, a book of these trades. Most people calling themselves dispersion PMs these days are mostly doing selective dispersion ie long single name vol short index and rarely look at the correlation edge alone (of course, beta can be thought as a correlation proxy).
I don't get this. My obligation is to buy the underlying at the strike price and I have the money for that (100%); it's like buying the underlying and selling covered calls.