Serving the Quantitative Finance Community

 
User avatar
marcodt
Topic Author
Posts: 0
Joined: May 24th, 2007, 2:39 pm

help on a project topic

May 24th, 2007, 8:15 pm

I recently received a external project, whose topic is:"Options on New Generation Correlation Products", and it's objective is to create a model to price options on correlation products(equity derivatives) . It's quite vague.I am working on a proposal now, and my current idea is just limited to what kind of correlation products(basket option with stock as underlying, or incoporated with stock index as well) I can write about, and maybe furthermore I can use MC and copula to model the price of specific products. However, if the project is only about this, it seems really not interesting and challenging.My question is1. Can the basket option be called New Generation correlation product, or I have misunderstood it. What is a so-called new generation corr product?2. It seems impossible to come up with a new pricing model, and could some one suggest any papers on the existing model(used in the market) for pricing.I am really appreciated that if someone could suggest something I can write about for it.
 
User avatar
marcodt
Topic Author
Posts: 0
Joined: May 24th, 2007, 2:39 pm

help on a project topic

May 25th, 2007, 7:47 am

Thanks outrun.So the hybrid products are quite new generation(started to be traded in the recent years)? Can U suggest some classical papers on hybrid product pricing models.
 
User avatar
marcodt
Topic Author
Posts: 0
Joined: May 24th, 2007, 2:39 pm

help on a project topic

May 25th, 2007, 12:27 pm

Currently, I am looking at the Correlation Swap. Is there any other products other than this.
 
User avatar
marcodt
Topic Author
Posts: 0
Joined: May 24th, 2007, 2:39 pm

help on a project topic

May 25th, 2007, 12:27 pm

Currently, I am looking at the Correlation Swap. Is there any other products other than this.
 
User avatar
Speedy
Posts: 0
Joined: May 16th, 2007, 1:04 pm

help on a project topic

May 30th, 2007, 5:44 pm

Covariance swaps - they work like correlation swaps, but you don't divide by the realized variances.