Implied Correlation: Var(a/c) Formula Explained

  • Context: Graduate 
  • Thread starter Thread starter volplayer
  • Start date Start date
  • Tags Tags
    Correlation
Click For Summary
SUMMARY

The discussion centers on the mathematical derivation of implied correlation based on implied volatility for FX options, specifically using the formulas Var(b/c) and Corr(a/c,a/b). The key formulas presented include Var(b/c) = Var(a/c) + Var(a/b) - 2*Sigma(a/c)*Sigma(a/b)*Corr(a/c,a/b) and its rearrangement for Corr(a/c,a/b). The participant seeks clarification on whether the Var(a/c) formula can be expressed with a plus sign instead of a minus sign in the correlation term, which is confirmed to be incorrect. The discussion emphasizes the importance of correctly identifying random variables in the context of these calculations.

PREREQUISITES
  • Understanding of implied volatility in FX options
  • Familiarity with variance and correlation in statistics
  • Knowledge of mathematical notation and random variables
  • Basic principles of the law of cosines
NEXT STEPS
  • Study the derivation of implied volatility models in FX options
  • Learn about the law of cosines and its applications in statistics
  • Explore advanced statistical concepts related to variance and correlation
  • Review the properties of random variables and their implications in financial mathematics
USEFUL FOR

Mathematicians, financial analysts, and anyone involved in quantitative finance who seeks to deepen their understanding of implied correlation and volatility in FX options.

volplayer
Messages
1
Reaction score
0
Hi,

Haven't studied math for a while and thought I'd ask you for help. It regards implied correlation based on implied volatility for FX options.

(b/c)=(a/c)/(a/b)

Var(b/c) = Var(a/c)+Var(a/b) - 2*Sigma(a/c)*Sigma(a/b)*Corr(a/c,a/b)

When breaking out the Corr(a/c,a/b) from the formula, we get the following:

Corr(a/c,a/b) = (Var(a/c)+Var(a/b)-Var(b/c)) / (2*Sigma(a/c)*Sigma(a/b))

Now let's break out (a/c)

(a/c) = (b/c) * (a/b)

Now I have understood that the Corr(b/c,a/b) formula is the following

Corr(b/c,a/b) = (Var(a/c)-Var(b/c)-Var(a/b)) / (2*Sigma(b/c)*Sigma(a/b))

Does this mean the Var(a/c) formula is like the following

Var(a/c) = Var(b/c)+Var(a/b) + 2*Sigma(b/c)*Sigma(a/b)*Corr(b/c,a/b) ?

I.e. you have a PLUS instead of a MINUS infront of the 2*Sigma*Sigma*Corr part?


Happy if someone could answer this.
 
Physics news on Phys.org
The minus sign is correct. What we have here is just a version of the law of cosines although your notation as fractions is very confusing. Try to determine your random variables in a proper way first.
 

Similar threads

  • · Replies 58 ·
2
Replies
58
Views
7K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 16 ·
Replies
16
Views
12K
  • · Replies 43 ·
2
Replies
43
Views
6K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 178 ·
6
Replies
178
Views
10K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K