MATH0085 - Asset Pricing in Continuous Time
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
MATH0085 - Asset Pricing in Continuous Time
Problem 1. Let (Ω ; F; {Ft }t
0 ; P) be some probability space and {Wt }t
0 and {Wˆt }t
0 be two (P; {Ft }t
0)-Brownian motions, where EP [(Wt − Ws )(Wˆt − Wˆs )] =
(t − s) for some
∈ (0; 1).
(a) [8 points] [SS] Let {Yt }t
0 and {Zt }t
0 be Ito processes governed by
Yt = Y0 +
as ds +
bs dWs ; and Zt = Z0 +
cs ds +
ds dWˆs :
If St = sin(t) + Yt2 + exp(Zt ) for all t ≥ 0, governed by the stochastic differential equation dSt = Kt dt + Lt dWt + Nt dWˆt , what are Kt , Lt and Nt in terms of Yt , Zt , at , bt , ct and dt for all t ≥ 0?
(b) [8 points] [S]
From above, if Xt = Yt Zt for all t ≥ 0 such that dXt =
t dt +
t dMt , where {Mt }t
0 is a (P; {Ft }t
0)-Brownian motion, what are
t and
t in terms of Yt , Zt , at , bt , ct and dt for all t ≥ 0?
(c) [9 points] [SS]
Define {Pt }0
t<
by Pt = sin(Wt )= cos(Wt ), where
= inf{t ≥ 0 : |Wt | = ![]()
}. Show that the following holds:
Pt =
Ps (1 + P ) ds +s(2)
(1 + Ps(2)) dWs :
Problem 2.
(a) [8 points] [SS]
Let (Ω ; F; {Ft }t
0 ; P) be some probability space and {Wt }t
0 be a (P; {Ft }t
0)- Brownian motion. For a finite
> 0, let {Ft }t
0 , {Gt }t
0 and {Ht }t
0 be
Ft = cosh(
Wt ) exp (
)
Gt =
+ Wt3 − 3tWt + Wt2 − t
Ht =
Wt
for all t ≥ 0, respectively. For each above, prove whether it is a (P; {Ft }t
0)- martingale or not.
(b) [8 points] [S]
Let {Bt }t
0 given by Bt = ert model the money-market account for some finite r ≥ 0 and let {Zt }t
0 be governed by dZt = −
t Zt dWt where {
t }t
0 is the market price of risk. What is the stochastic differential equation of {
t }t
0 given by
t = Zt =Bt ? Also, prove that {
t }t
0 is a (P; {Ft }t
0)- supermartingale.
(c) [9 points] [U]
Let
0 be some finite constant an {Et }0
t<
be governed by the following:
Et =
+
Es(3) ds −
Es(2) dWs ; E0 =
:
Provide an explicit solution to the stochastic integral above – that is, what is the function f if the solution is Et = f (t; Wt ). Based on this solution, what should be the definition of
in terms of Wt and
?
Problem 3. Let (Ω ; F; {Ft }t
0 ; P) be some probability space and {Wt }t
0 be a (P; {Ft }t
0)-Brownian motion. Let {Bt }t
0 governed by dBt = rBt dt model the money-market account, where B0 = 1 and r > 0 is finite. Let {Xt }t
0 model asset price dynamics where dXt =
Xt dt +
Xt dWt for some finite constants
∈ R and
> 0 with X0 = x.
(a) [8 points] [S]
Prove that {Zt }t
0 defined by
Zt =
exp ( − ![]()
2 t −
Wt )
is a (P; {Ft }t
0)-martingale if
= (
− r)=
.
(b) [8 points] [SS]
Define the risk-neutral probability measure Q as follows:
FT = exp ( − ![]()
2 T −
WT ) :
For some
> 0 and K ≥ 0, derive the following option price: V0 = e rTEQ [(X
− K)+]. (Hint: These are called power options)
(c) [9 points] [SS]
Now define the probability measure U as follows:
dU XT
dQ FT xBT
Show that for any 0 ≤ s ≤ t ≤ T , the following holds:
EU [
Fs] =
Problem 4. Let (Ω ; F; {Ft }t
0 ; Q) be a probability space where Q is the risk- neutral probability measure and {Wt }t
0 is a (Q; {Ft }t
0)-Brownian motion. For interest rate r > 0 and volatility
> 0 (both finite), let {Xt }t
0 be governed by dXt = rXt dt +
Xt dWt .
(a) [9 points] [U] If g(x) is a non-negative convex function for x ≥ 0 with g(0) = 0, prove that {St }t
0 given by St = e rtg(Xt ) is a (Q; {Ft }t
0)- submartingale. (Hint: Since g is convex, g ((1 −
)x1 +
x2 ) ≤ (1−
)g(x1 )+
g(x2 ) for 0 ≤
≤ 1 and 0 ≤ x1 ≤ x2 . Set x1 = 0 and x2 = x)
(b) [8 points] [S]
Define the stochastic integral process {Mt }t
0 as follows:
Mt =
Wt3 −
Ws ds:
Compute the expected value EQ [Mt ] and the variance VarQ [Mt ].
(c) [8 points] [SS]
Let {Wˆt }t
0 be another (Q; {Ft }t
0)-Brownian motion which is independent of {Wt }t
0 . Define Yt =
Wt=
2 for some
> 0 and
t = tWˆ1=t for t ≥ 0, with Y0 =
0 = 0. If Rt = Yt =
t whenever
t
0, what is the distribution Q(R1 ∈ dr)?
2022-04-27