FRM Part II · FRM Exam Part II · Financial Correlation Modeling - Bottom-Up Approaches
Two independent standard normal variables Z1 and Z2 are drawn in a Monte Carlo simulation. To simulate correlated shocks with correlation 0.60, an analyst sets X1 = Z1 and X2 = 0.60 Z1 + k Z2. What value of k makes X2 have unit variance?
The value is 0.80. For X2 to have unit variance, 0.60 squared plus k squared must equal one, so k squared is 0.64 and k is 0.80. This is the Cholesky-style construction, and it also preserves correlation of 0.60 with X1.
- A0.40
- B0.64
- C0.80Correct
- D0.84
Explanation
Var(X2) = 0.36 + k^2 = 1, so k^2 = 0.64 and k = 0.80. Check: Corr(X1,X2) = 0.60 because Cov = 0.60. Using 0.40 (1 - rho) or 0.64 (k^2 without the root) are common mistakes.
Did you get it right without looking?
One question tells you little. A timed set on Financial Correlation Modeling - Bottom-Up Approaches shows your real accuracy, how long you take and where you lose marks.
More Financial Correlation Modeling - Bottom-Up Approaches questions
- A risk analyst reviews a synthetic CDO referencing 100 equally weighted credit default swaps. The analyst raises the assumed pairwise defaul…
- A risk analyst models the joint behavior of two assets by letting each asset follow its own geometric Brownian motion, with the Brownian inc…
- A risk analyst is explaining why a bank's models for correlated defaults differ from the models used for correlated equity returns. Which st…
- A risk analyst at a bank models the joint default behavior of 100 corporate borrowers using a one-factor Gaussian copula. Each borrower's ma…
- A bank prices a CDO tranche using a Gaussian copula. When the implied correlation needed to match the market price of the equity tranche dif…
- Under a one-year horizon, two firms A and B each have a cumulative default probability of 2.275%, so the corresponding standard normal thres…