FRM Part II · FRM Exam Part II · Regression Hedging and Principal Component Analysis
A single-variable regression hedge regresses daily changes in the yield of a bond being hedged (dependent variable) on daily changes in the yield of a hedging instrument. The estimated beta is 1.20. The bond's DV01 is USD 12,000 and the hedging instrument has a DV01 of USD 400 per USD 1 million face. What face amount of the hedging instrument should be sold, ignoring the regression's error term?
Sell USD 36 million face of the hedging instrument. The naive DV01 hedge is 12,000 divided by 400, or 30 million, and it is scaled up by the regression beta of 1.20 because the bond's yield moves 1.2 times as much as the hedge's yield.
- AUSD 30 million
- BUSD 25 million
- CUSD 36 millionCorrect
- DUSD 21 million
Explanation
Regression hedge: F = (DV01 of bond / DV01 of hedge) x beta = (12,000/400) x 1.20 = 30 x 1.20 = 36 million. Ignoring beta gives 30 million, the naive DV01 hedge. Dividing by beta gives 25 million, a common inversion error.
Did you get it right without looking?
One question tells you little. A timed set on Regression Hedging and Principal Component Analysis shows your real accuracy, how long you take and where you lose marks.
More Regression Hedging and Principal Component Analysis questions
- A trader holds a long position in a bond with a DV01 of USD 4,500 per basis point. She wants to hedge against a parallel shift in yields usi…
- A bank hedges a bond portfolio's term structure risk using the first three principal components of yield changes, which together explain 97%…
- A portfolio manager hedges a bond position using a PCA-based approach with the level and slope factors. The bond has a level factor exposure…
- PCA on a covariance matrix of three interest rate changes gives eigenvalues of 8.0, 1.5 and 0.5 (in bp² units). What proportion of total var…
- An analyst runs PCA on the covariance matrix of five key rate changes. Eigenvalues are 9.0, 3.0, 1.2, 0.5 and 0.3. A portfolio's risk is mod…
- A risk manager runs principal component analysis on daily changes in the 2-, 5-, 10- and 30-year par swap rates. The first three components …