Additional questions for chapter 4

1. A stock price is currently $ 100. Over the next two six-month periods it is expected
to go up by 10% or go down by 10%. The risk-free interest rate is 8% per annum
with continuous compounding.

(i) What is the value of a one-year European call option with a strike price of $
100.
(ii) What is the value of a one-year European put option with a strike price of $
100.
(iii) Verify that the European call and the European put satisfy put-call parity.

Solution:
Parameters are u = 0.1, d = −0.1, 1 + r = e0.5×0.08 . So the risk-neutral probability is
p∗ = 0.7. After evaluation of the options at the terminal nodes we use the risk-neutral
valuation to get (i)
£ ¤
πC (0) = e−2(0.5×0.08) 0.72 × 21 + 2 × 0.7(1 − 0.7) × 0 + (1 − 0.7)2 × 0 = 9.61

and (ii)
£ ¤
πP (0) = e−2(0.5×0.08) 0.72 × 0 + 2 × 0.7(1 − 0.7) × 1 + (1 − 0.7)2 × 19 = 1.92

(iii) For put-call parity one has to verify S − πC + πP = Ke−r , here :

100 − 9.61 + 1.92 = 100e−0.08 .

2. Assume a standard 3-period CRR binomial model. The price of the stock is currently
$100. The risk-free interest rate with continuous compounding is 6% per annum.
Over the next three 4 month periods, the stock is expected to go up by 8% or go down
by 7% in each period.

(a) What is the value of a one-year European call with strike price $103?
(b) What is the value of a one-year European put with strike price $103?
(c) Verify the Put-Call parity for the European call and the European put.

Solution:
We first calculate the Martingale probability in the tree. We get

r−d e0.06/3 − 1 + 0.07

p= = = 0.6013423
u−d 0.08 + 0.07
(a) The tree for the call option looks as follows:

125.97
116.64 ©©© 22.97
H
108 ©©© 15.679 HH
108.475
´ H
S=100 ´ 10.503 HH 5.475
´ 100.44 ©©
©
C=6.93 QQ 3.2272 HH
Q 93 ©© H 93.409
©
1.9022 H
HH 0
86.49 ©©©
H
0 HH 80.435
0

time t = 0 t = 1/3 t = 2/3 t=1

(b) The tree for the put option is:

125.97
116.64 ©©© 0
HH
108 © ©© 0 H 108.475
S=100 ´
´ 1.46448 HHH 0
´ 100.44 ©©©
C=3.936QQ 3.7477 HH
Q 93 ©© H 93.409
©
7.8635 HH 9.5908
H 86.49
©©©
HH
14.47 H 80.435
22.5643

time t = 0 t = 1/3 t = 2/3 t=1

(c) The Put-Call parity holds:

C −P = 6.9342−3.936 = 2.9982 = 100−103e−0.06 = 100−97.0017 = S −Ke−rT .

3. Consider a 3-period Cox-Ross-Rubinstein model. The annual interest rate is r = 0.05
(discrete), u = 0.1 and d = −0.1. The initial price of the stock is S(0) = 100. The
time horizon is T = 3 years.
(a) Calculate the risk-neutral probability and the stock prices at each node in the
binomial tree (correct up to 2 decimal places after the decimal point).
(b) Calculate the value of the European option with payoff

 sup St − ST St < 110 ∀t
P (T ) = 0≤t≤T
0 otherwise

(c) Find a replicating portfolio for the above option for the first trading period.
Solution:

r−d
(a) For the risk-neutral probability we get p = u−d
= 34 . The tree with the stock
prices and the value of the option is
133.1
½ 0
121 ½
½
­­ 0
Z
Z
Z 108.9
­
110 ­ 0
­
£ 0 J
£ J 108.9
£ ½
£
J
99 ½ 0
JJ ½
£
£ 0 Z
Z
Z 89.1
£
S=100 £ 0
£
P =1.25 B
B 108.9
B ½ 0
B 99 ½
½
­­ 2.595
B Z
B Z
Z 89.1
B ­
90 ­ 10.9
B
B ­
5.24 J 89.1
J
J ½ 10.9
JJ 81 ½
½
14.23 Z
Z
Z 72.9
27.1

time t = 0 t = 1/3 t = 2/3 t=1

(b) The replicating portfolio can be found by solving the equations
1.05 · ϕ1 + 110 · ϕ2 = 0
1.05 · ϕ1 + 90 · ϕ2 = 5.24

As solution we get ϕ1 = 27.45 and ϕ2 = −0.262.

4. Construct a three period binomial tree using the parameters r = 0.1 (discrete, per
period), u = 0.15, d = −0.05 and S0 = 100.

(a) Find the price of a European Put P with strike 105 and maturity date T = 3.
(b) Find the price of the knock in Call option C with knock in level H = 110, strike
K = 90 and maturity date T = 3, i.e.
(
(S(T ) − 90)+ ∃t : St > H = 110
C=
0 St ≤ H = 110 ∀t.

Solution:
The risk neutral probability is
r−d 0.1 + 0.05 3
p= = =
u−d 0.15 + 0.05 4

We first set up a tree with the stock price movements, then compute the values of
the two options:
152.09
132.25 ½
0
½ 62.09
0 ½
­­ 50.43
Z
Z
125.64
­ Z 0
115 ­
0.0625 ­ 35.64
£ J 125.64
£ 40.62 J
£ 109.25 ½
0
J
£ JJ 0.275 ½ 35.64
£ ½
£
Z 103.79
27.43 Z
Z
S=100 £
£ 1.21
P =0.345£ 13.79
B 125.64
C=31.46 B
109.25 ½
0
B
B ½ 35.64
0.275 ½
B Z 103.79
B ­­ 24.3 Z
­ Z 1.21
B 95 ­
B
B 1.37 ­ 0
J 103.79
16.57 J
90.25 ½
1.21
J
JJ 5.2 ½ 0
½
Z 85.74
0 Z
Z 19.26
0

time t = 0 t = 1/3 t = 2/3 t=1

5. Assume a 3-period Cox-Ross-Rubinstein model. The annual interest rate with con-
tinuous compounding is r = 0.06. The volatility of the stock is σ = 0.2 with a price
of S(0) = 100. Furthermore, there exists an American Put with maturity date T = 1
und strike K = 90.
(a) Calculate the risk-neutral probability and the stock prices at each node in the
binomial tree (correct up to 2 decimal places after the decimal point).
(b) Calculate the value of the American Put for all nodes in the tree.
(c) What is the optimal stopping time? Justify your answer.
Solution:

(a) The parameter-values are

1 √ √
∆ = , 1+rd = er∆ = 1.0202, 1+u = eσ ∆ = 1.1224, 1+d = e−σ ∆
= 0.8909.
3
For the risk-neutral probability we get
rd − d
p∗ = = 0.5584.
u−d
The tree with the stock prices is
©© 141.40
125.98 ©
H
©© HH
112.24 ©
´ HH 112.24
´ H ©
S=100 ´
Q 89.09 ©©
HH
Q
Q 89.09 ©© H
©
H 89.09
HH ©©
79.38 ©
H
HH
70.72
time t = 0 t = 1/3 t = 2/3 t=1
(b) The prices for the american Put are
©© 0
0 ©
HH
© ©© H
´ 0.17 HH 0
´ H ©©
P=2.176´
Q 0.394 ©
HH
Q
Q 4.812 ©© H
©
HH 0.91
H ©©
max{10.62, 8.84} ©
HH
H 19.28
time t = 0 t = 1/3 t = 2/3 t=1
(c) Let Ω = {u, d}3 . The optimal exercise date is
(
n = 2 ω ∈ {ddu, ddd}
τ (ω) =
n = 3 otherwise.
1
For ω ∈ {(ddu), (ddd)}, we have 1+r d
E[p∗ f32 + (1 − p∗ )f33 ] < (K − S0 (1 + d)2 )+ .
Here fij denotes the price of the claim in period i with j-down movements.
6. Assume that we have a three period CRR model with initial stock price S = $150,
interest rate r = 0.05 and volatility σ = 0.2.

(a) What is the value of an American Put with strike $150, which matures in 6
months?
(b) What is the value of an American Call with strike $150, which matures in 6
months?
(c) Verify that the following inequalities hold:

S − K ≤ CA − PA ≤ S − Ke−rT

Solution:
The martingale probability is p = 0.5308 with u = 0.085, d = −0.0784 and r =
0.0084.
(a), (b) For the American Put and Call we get:
191.63
0
176.61 ¡
¡ 41.63
0 ¡
162.76 ¡ @ 162.76
¡ 27.85 @
2.55 ¡ @ 0
S=150 ¡ @ 150 ¡
¡ 17.78 @ ¡ 12.76
PA =7.57 ¡ @ 5.47 ¡
CA =11.01 @@ 138.24 ¡
¡
6.72 @
@
138.24
@ 13.40 ¡ @ 11.76
@ 127.40 ¡
3.54 @ ¡ 0
@ 22.60 ¡
@ 117.41
0 @
@ 32.59
0

time t = 0 t = 1/6 t = 2/6 t = 0.5

(c) We have 0 ≤ 11.01 − 7.57 = 3.44 ≤ 3.7035 = 150 − 150 · e−0.025

7. Show that a security market is arbitrage-free with respect to Φ iff it is arbitrage-free
with respect to Φa . Here Φ is the set of all self-financing trading strategies and Φa
is the set of all admissible strategies, that means all ϕ ∈ Φ with Vϕ (t) ≥ 0 t =
0, . . . , T .
Solution:
First note, if ϕ ∈ Φa is an arbitrage strategy, then it is by definition of Φa also a
strategy in Φ. We now have to show that if we have an arbitrage strategy ϕ ∈ Φ,
then there exists an arbitrage strategy ψ ∈ Φa .
Assume that ϕ ∈ Φ is an arbitrage strategy. Then we have Vϕ (0) = 0, P (Vϕ (T ) ≥
0) = 1 and P (Vϕ (T ) > 0) > 0. We have to distinguish between two cases:

Case 1: Vϕ (t) ≥ 0 t = 0, . . . , T . Then ϕ ∈ Φa and we found the admissible

arbitrage strategy.
Case 2: ∃t∗ , A ∈ Ft∗ with Vϕ (t∗ , ω) < 0 ∀ω ∈ A and Vϕ (t) ≥ 0 t > t∗ . Then define
a new strategy ψ. Set ψ(u, ω) = 0 ∀ω ∈ Ac ∀u. Furthermore ψ(u, ω) = 0 ω ∈
A and u ≤ t∗ . For the remaining possibilities set

Vϕ (t∗ , ω)
ψ0 (u, ω) = ϕ0 (u, ω) − ∀ω ∈ A u > t∗
S0 (t∗ , ω)

and
ψi (u, ω) = ϕi (u, ω) ∀ω ∈ A i = 1, . . . , d u > t∗
We have to show that this strategy is self-financing and admissible. For ω ∈ Ac
we clearly have no problem. There is nothing to show for ω ∈ A, u ≤ t∗ . ψ is
also clearly self-financing for u > t∗ + 1 as it just replicates the other strategy
there. We have to show that ψ(t∗ )S(t∗ ) = ψ(t∗ + 1)S(t∗ ). For ω ∈ A we have

ψ0 (t∗ + 1)S0 (t∗ ) = ϕ0 (t∗ + 1)S0 (t∗ ) − Vϕ (t∗ ) and ψi (t∗ + 1) = ϕi (t∗ + 1)

Thus we get

ψ(t∗ +1)S(t∗ ) = 1A (ϕ(t∗ +1)S(t∗ )−Vϕ (t∗ )) = 1A (ϕ(t∗ )S(t∗ )−Vϕ (t∗ )) = 0 = ψ(t∗ )S(t∗ )

It remains to show that ψ is admissible and an arbitrage opportunity. We get

Vψ (t) = 0 t ≤ t∗

and
µ ¶ µ ¶
∗ S0 (t) ∗ S0 (t)
Vψ (t) = 1A ϕ(t)S(t) − Vϕ (t ) = 1A Vϕ (t) − Vϕ (t ) ≥0
S0 (t∗ ) S0 (t∗ )

and > 0 on A for t = T because Vϕ (t∗ ) < 0. We also have that Vψ (t) = 0 ∀t ≤ t∗ .
Therefore ψ is admissible and an arbitrage opportunity.
8. (a) State the Black-Scholes formula for an European Call and Put. (Hint: The
Put-Call parity C − P = S − Ke−r(T −t) might be useful)
(b) Replicate the European straddle with payoff D(T ) = |S(T ) − K| using standard
European options.
(c) What is the Black-Scholes price of the straddle?
(d) What is the ∆ of the straddle? How much does the value of the straddle approx-
imately change if the stock price changes from St to St + ε? (Hint: The ∆ of
the Call is N (d1 ))
Solution:

(a) The Black-Scholes formula for an European Call and Put is

C(t) = S(t)N (d1 ) − Ke−r(T −t) N (d2 )

P (t) = Ke−r(T −t) N (−d2 ) − S(t)N (−d1 )

where
³ 2
´
log(S/K) + r + σ2 (T − t)
d1 = √
σ T −t
√
d2 = d1 − σ T − t.

(b) We can replicate the straddle D(T ) = |S(T ) − K| by buying one call and one
put, both with strike K.
(c) The Black-Scholes price of the straddle is

D(t) = C(t) + P (t) = S(t)N (d1 ) − Ke−r(T −t) N (d2 ) + Ke−r(Tt ) N (−d2 ) − S(t)N (−d1 ) =
= S(t)(2N (d1 ) − 1) − Ke−r(T −t) (2N (d2 ) − 1).

(d) The Delta of the straddle is

∆D = ∆C + ∆P = N (d1 ) − N (−d1 ) = 2N (d1 ) − 1.

When the stock price changes from St to St + ε, then the price of the straddle
changes about ε(2N (d1 ) − 1).
9. Consider a financial market in which the Black-Scholes formula for a European call
option holds. The risk-free interest rate (cont. compounding) is r. The underlying
stock has value S with volatility σ. For a European call with strike K and maturity
T, show that the following relations hold:
∂C
∆= = N (d1 )
∂S
∂C N 0 (d1 )
Γ= = √
∂S 2 Sσ T − t
∂C SN 0 (d1 )σ
Θ= =− √ − rKe−r(T −t) N (d2 )
∂t 2 T −t
∂C
ρ= = K(T − t)e−r(T −t) N (d2 )
∂r
∂C √
ν= = SN 0 (d1 ) T − t
∂σ
Show that the call satisfies the partial differential equation
∂C ∂C 1 2 2 ∂ 2 C
+ rS + σ S − rC = 0.
∂t ∂S 2 ∂S 2

Solution:
We first show that SN 0 (d1 ) = Ke−r(T −t) N 0 (d2 ):

0 −r(T −t) 0 1 ³ −d21 /2 −r(T −t) −d22 /2

´
SN (d1 ) − Ke N (d2 ) = √ Se − Ke e =
2π
1 ³ −d21 /2 √
−r(T −t) −d21 /2+d1 σ T −t−σ 2 (T −t)/2
´
= √ Se − Ke e =
2π
µ ¶
1 −d21 /2 −d21 /2 K −r(T −t) d1 σ T −t−σ 2 (T −t)/2
√
=√ Se − Se e e =
2π S
µ ¶
1 −d21 /2 −d21 /2 K −r(T −t) log(S/K)+(r+σ 2 /2)(T −t)−σ 2 (T −t)/2
=√ Se − Se e e =
2π S
1 ³ −d21 /2 −d21 /2
´
=√ Se − Se =0
2π

Now we calculate the Greeks:

(a)
∂C ∂d1 ∂d2
∆= = N (d1 ) + SN 0 (d1 ) − Ke−r(T −t) N 0 (d2 ) =
∂S µ ∂S ¶ ∂S
∂d1 ∂d2
= N (d1 ) + SN 0 (d1 ) − =
∂S ∂S
= N (d1 )

(b)
∂C ∂∆ 0 ∂d1 N 0 (d1 )
Γ= = = N (d1 ) = √
∂S 2 ∂S ∂S Sσ T − t
 (c)

∂C ∂d1 ∂d2
Θ= = SN 0 (d1 ) − Ke−r(T −t) N 0 (d2 ) − Kre−r(T −t) N (d2 ) =
∂t µ ∂t ¶ ∂t
0 ∂d1 ∂d2
= SN (d1 ) − − Kre−r(T −t) N (d2 ) =
∂t ∂t
σ
= −SN 0 (d1 ) √ − rKe−r(T −t) N (d2 )
2 T −t

(d)

∂C ∂d1 ∂d2
ρ= = SN 0 (d1 ) − Ke−r(T −t) N 0 (d2 ) + K(T − t)e−r(T −t) N (d2 ) =
∂r µ ∂r ¶ ∂r
0 ∂d1 ∂d2
= SN (d1 ) − + K(T − t)e−r(T −t) N (d2 ) =
∂r ∂r
= K(T − t)e−r(T −t) N (d2 )

(e)

∂C ∂d1 ∂d2
ν= = SN 0 (d1 ) − Ke−r(T −t) N 0 (d2 ) =
∂σ µ ∂σ ¶ ∂σ
∂d1 ∂d2
= SN 0 (d1 ) − =
∂σ ∂σ
√
= SN 0 (d1 ) T − t.

The partial differential equation holds because:

∂C ∂C 1 2 2 ∂ 2 C
+ rS + σ S − rC =
∂t ∂S 2 ∂S 2
σ
= −SN 0 (d1 ) √ − rKe−r(T −t) N (d2 )+
2 T −t
+ rSN (d1 )+
1 N 0 (d1 )
+ σ2S 2 √ +
2 Sσ T − t
+ rC =
= r(SN (d1 ) − Ke−r(T −t) N (d2 ) − C) = 0.
10. Prove the following limit relations used in the proof of Proposition 3.5.1, assuming
that kn → ∞ (n → ∞):
1 p ³r σ´
lim p̂n = , lim kn (1 − 2p̂n ) ∆n = −T +
n→∞ 2 n→∞ σ 2

Solution:
We have the following definitions for the variables:
T
∆n =
kn
√
un = eσ ∆n − 1
√
dn = e−σ ∆n
−1
rn = er∆n − 1
rn − dn
p∗n =
un − dn
1 + un
p̂n = p∗n
1 + rn

Then, we get for the first limit relation:

√
1 + un er∆n − e−σ ∆n 1
lim p̂n = lim p∗n lim = lim σ√∆ √ = .
n→∞ n→∞ n→∞ 1 + rn n→∞ e n − e−σ ∆n 2
| {z }
→1(n→∞)

In order to show the last equality, it suffices to show that

2
erx − e−σx 1
lim = .
x→0+ eσ x − e−σx 2
√
as ∆n → 0+ (n → ∞).
By L’Hospital we get
2 2
erx − e−σx 2xrerx + σe−σx 1
lim == lim = .
x→0+ eσ x − e−σx x→0+ σeσx + σe−σx 2

For the second limit relation we get:

à √ !
p p er∆n − e−σ ∆n e−r∆n
lim kn (1 − 2p̂n ) ∆n = lim T kn 1− 2 σ √∆ √ √ =
n→∞ n→∞ e n − e−σ ∆n e−σ ∆n
√ √
√ 1 − e−2σ ∆n − 2 + 2e−σ ∆n −r∆n
= lim T ¡ √ ¢
n→∞ √1 1 − e−2σ ∆n
kn
√ ³ √ ³σ r ´´ ³σ r´
= T − T + = −T + .
2 σ 2 σ

For the second to last equation it suffices to show that:

−e−2σx − 1 + 2e−σx−rx
2
√ ³σ r´
lim =− T +
x→0+ √x (1 − e−2σx ) 2 σ
T
 √
as ∆n → 0+ (n → ∞).
We are using L’Hospital twice and get:
2
−e−2σx − 1 + 2e−σx−rx
lim =
x→0+ √x (1 − e−2σx )
T
√ 2σe−2σx − 2(σ + 2rx)e−σx−rx2
= lim T =
x→0+ (1 − e−2σx ) + 2xσe−2σx
√ −4σ 2 e−2σx + 2(σ + 2rx)2 e−σx−rx2 − 4re−σx−rx2
= lim T =
x→0+ 2σe−2σx + 2σe−2σx − 4xσ 2 e−2σx
√ −4σ 2 + 2σ 2 − 4r √ ³σ r´
= T =− T + .
4σ 2 σ
http://www.springer.com/978-1-85233-458-1