For the two-factor CIR model proposed by Longstaff and Schwartz (1992), the short rate r(t) is defined

Question:

For the two-factor CIR model proposed by Longstaff and Schwartz (1992), the short rate r(t) is defined by 

r = ax + By,

where α and β are positive constants, and α = β. Under the risk neutral measure, the risk factors x and y are governed by

dx = (y - 8x) dt + xdZ dy (ny) dt + ydZ2, =where Z1 and Z2 are uncorrelated standard Brownian processes. Let V denote the instantaneous variance of changes in the short rate. 

(a) Show that

V = ax + By.

(b) Using Ito’s lemma, show that the dynamics of r and V are governed by 

dr = (  +  dV = 20 + -       r  V V (  ) =artin-  +  ta? -   + dZ1 + , (  )    Br - V (  )   dZ1 + 2, -  V dt

Note that r and V together form a joint Markov process. 

(c) Show that r has a long-run stationary (unconditional) distribution with mean

and variance E[r] var(r) =   +     + 282 22Similarly, show that V also has a stationary distribution with mean

and variance ELVI-/+ E[V] = var(V) = 'n    + 282 22

(d) Let B(r,V,τ) denote the price function of a unit discount bond with τ periods until maturity. Show that

where and B(r, V, t) = AY (T) Bn (T) exp(KT + C(t)r + D(t)V), A(T) = B(T) = C(T) = D(T) = 20 (8 + 0) (exp(pt)(e) Suppose a < , show that V (t) is limited to the range (ar (t), Br(t)) at any time t.

Fantastic news! We've Found the answer you've been seeking!

Step by Step Answer:

Related Book For  answer-question
Question Posted: