Page 1 of 1
implied correlation
Posted: November 26th, 2008, 7:02 am
by chtebel
From What kind of product it would be possible to extract implied corraletion between two products. and what would be the procedure?I heard it was possible but i don't find anything on it. (The only thing i found is how to calculate the implied correlation (a scalar) of an index but it doesn't seem really useful for that problem)1°)if the actions are containing in the same index citigroup/microsoft2°) and if they are not. ex: toyota/Citigroupthx a lot
implied correlation
Posted: November 26th, 2008, 8:38 am
by phil451
Implied Correlation is derived from the price of a CDO. There are a couple of methodologies for doing this. Check the internet.If you are familiar with options Implied Correlation is analogous to implied volatility, so you get an Implied Correlation skew similar in nature to the implied volatility skew
implied correlation
Posted: November 26th, 2008, 8:58 am
by daveangel
1. Index variance versus constituent stock variance aka dispersion trading2. implied vol from cross fx options versus the pairs
implied correlation
Posted: November 26th, 2008, 11:51 am
by chtebel
QuoteOriginally posted by: daveangel1. Index variance versus constituent stock variance aka dispersion trading2. implied vol from cross fx options versus the pairsthanks daveangelIt would be very nice if you could explain the procedure to extract implied correlation, especially for 2°) having implied vol from fx options is ok but what is the form of the calculus and the intuition behind?
implied correlation
Posted: November 26th, 2008, 12:24 pm
by daveangel
if you have the vol for exchange rate X1 and X2 (both against the same base) and the vol the cross (X12) then relationship is given by the cosine rule:vol12^2 = vol1^2 + vol2^2 + 2*rho*vol1*vol2hence rho = (vol12^2 - vol1^2 - vol2^2)/(2*vol1*vol2)
implied correlation
Posted: November 26th, 2008, 12:46 pm
by chtebel
QuoteOriginally posted by: daveangelif you have the vol for exchange rate X1 and X2 (both against the same base) and the vol the cross (X12) then relationship is given by the cosine rule:vol12^2 = vol1^2 + vol2^2 + 2*rho*vol1*vol2hence rho = (vol12^2 - vol1^2 - vol2^2)/(2*vol1*vol2)let me summarize and tell me if I m wrong you re saying that if you want to get the implied correlation between CitiG et Toyota for instance, you take the implied volatility of the EUR/USD (vol1) exchange rate, the implied volatility of the EUR/JPY (vol2) you calculate rho like you said: rho = (vol12^2 - vol1^2 - vol2^2)/(2*vol1*vol2) ?so you need vo12 but how to get vol12, which is the implied covariance of toyota and citig, from an option as a spread option? and if yes, so we must have an spread option existingbecause my purpose is to calculate an implied correlation matrix for basket pricing with international valuesthx
implied correlation
Posted: November 26th, 2008, 12:50 pm
by daveangel
no - sorry I am talking about figuring out the implied correla between EUR/USD and JPY/USD given EURJPY, EUR/USD and JOY/USD vols.if you have options on a basket of Citi and Toyota as well as options on C and Toyota then you can apply the same principle.
implied correlation
Posted: November 26th, 2008, 1:12 pm
by chtebel
ok. no problem.but it seems you need a basket option with citi and toyota to get an implied correlation. But I heard it xas possible without it. because it s hard to find it for all equitities.
implied correlation
Posted: November 26th, 2008, 1:21 pm
by daveangel
unfortunately there arent that many observable points for equities.
implied correlation
Posted: July 24th, 2013, 2:44 pm
by lp1
typo