FRM Part II · FRM Exam Part II · The Art of Term Structure Models: Volatility and Distribution
A manager must choose between a normal model with constant basis-point volatility and a CIR model for hedging a book when the short rate has historically moved from 8% to 1%. Empirically, basis-point volatility fell as rates fell. Which conclusion is best supported?
CIR fits the observation that basis-point volatility falls when rates fall, because its volatility is sigma times the square root of the rate. A normal model holds volatility constant and so would overstate volatility at low rates. The two models do not imply the same volatility.
- ACIR better captures the observed link between rate level and volatility, while the normal model would overstate volatility at low ratesCorrect
- BThe normal model is preferable because it makes volatility fall as rates fall
- CBoth models imply the same volatility at every rate level, so the choice is irrelevant
- DCIR would overstate volatility at low rates because volatility is proportional to the rate squared
Explanation
CIR volatility σ√r declines as r falls, matching the observed pattern. A normal model holds basis-point volatility constant, so at 1% it would overstate volatility relative to the data. CIR volatility is not proportional to r squared, and the models differ in their volatility dependence.
Did you get it right without looking?
One question tells you little. A timed set on The Art of Term Structure Models: Volatility and Distribution shows your real accuracy, how long you take and where you lose marks.
More The Art of Term Structure Models: Volatility and Distribution questions
- The Black-Karasinski model is commonly written d(ln r) = a(t)[ln θ(t) − ln r]dt + σ(t)dw. Which description is correct?
- A desk calibrates a model with time-dependent volatility σ(t) = σ0·e^(−λt) to the term structure of volatility, as in the Ho-Lee model with …
- In a Ho-Lee model with constant volatility σ = 1.20% per year, the short rate is currently 4.00%. Ignoring the drift's effect on dispersion,…
- In the Ho-Lee model, the short rate follows dr = λ(t)dt + σ dw with σ = 1.20% per year. A risk manager wants the standard deviation of the s…
- A desk calibrates the Hull-White model dr = k[θ(t) − r]dt + σ(t)dw. Compared with the Ho-Lee model with the same time-dependent drift, which…
- In Tuckman's treatment of term structure models, Model 3 is dr = λ(t)dt + σ dw, where the drift is time-dependent and the volatility is cons…