Modification of Black-Scholes Model: 7.1 Dividend
Modification of Black-Scholes Model: 7.1 Dividend
Modification of Black-Scholes Model: 7.1 Dividend
78
Chapter 7
Modification of Black-Scholes
model
7.1
7.1.1
Dividend
Continuous dividend yield model
Let q denote the constant continuous dividend yield which is known.1 In other
words, the holder receives a dividend, qSdt, within the time interval dt. The share
value is lowered after the dividend payout so that the expected rate of return
of a share becomes q by the continuous dividend yield q. The geometric
Brownian motion model becomes
dS
= ( q)dt + dZ
S
1
(7.1)
The continuous yield model is extremely useful to options on foreign currencies where the
continuous dividend yield can be considered as the yield due to the interest earned by the foreign
currency at the foreign interest rate rf . In the pricing model for a foreign currency call option,
we can simply set q = rf . S is the domestic currency price of a unit of foreign currency and the
exchange rate is assumed to follow the lognormal diffusion process. The model in this section
applied to the currency option, assumes that both the domestic and foreign interest rates are
constant.
79
By having number of the share, a share holder gains a dividend, (qdt)S. The
value of a portfolio
1
+
call option
underlying shares
c 2 2 c
c
S 2 2 + qS dt +
t
2
S
S
dS
(7.2)
c
S .
The portfolio is
then risk-free and should earn the risk-free interest rate, r. Thus
d =
c
c 2 2 c
S 2 2 + qS
t
2
S
S
c
dt = r c + S
S
dt
(7.3)
By rearranging the equation above, we have the modified form of the BlackScholes equation:
c 2 2 2 c
c
+ S
+ (r q)S
rc = 0
2
t
2
S
S
(7.4)
(7.5)
d1 = X
2
2 )
, d2 = d1
(7.6)
(7.7)
The put call parity can be derived using Eqs.(7.5) and (7.7),
p = c Seq + Xer ,
(7.8)
(7.9)
7.1.2
80
Discrete dividend
Suppose the underlying asset pays N discrete dividends at known payments dates
t1 , t2 , , tN of amounts D1 , D2 , , DN , respectively. If the actual amounts of
dividends and ex-dividend dates are known then we can assume that the asset
price is composed of two components:
risk-free component that will be used to pay the known dividends during
the life of the option.
risky component which follows a stochastic process.
The risk-free component is taken to be the present value of all future dividends
discounted at the risk-free interest rate. The value of the risky component St is
St = St D1 er1 D2 er2 DN erN for t < t1
St = St D2 er2 DN erN for t1 < t < t2
St = St
for t > tN
(7.10)
7.2
2
S
S
(7.11)
81
w
2 ( ) 2 w
2 ( ) w
=
+
r(
)
q(
)
2 y 2
2
y
r( )d ]c,
(7.12)
{y +
1
(y, ) = q R
exp
2
2 0 (u)du
[r(u) q(u)
R
2 0 2 (u)du
2 (u)
2
2 ]du}
(7.13)
w(y, ) =
w(, 0)(y , )d
(7.14)
c = SN (d1 ) exp
r(u)du
(7.15)
where
d1 =
1
qR
d2 = d1
7.3
2 (u)du
sZ
S
ln
+
X
"
2 (u)
r(u) q(u) +
du
2
2 (u)du.
(7.16)
Transaction costs
There have been continual efforts to construct hedging strategies that best replicate the payoffs of derivative securities in the presence of transaction costs. The
transaction cost occurs in buying and selling. Suppose that the transaction cost
is proportional to the amount of money involved in the transaction and the rate
of proportion is k2 . For buying (+) or selling (-) of || shares at the price S, the
transaction cost is k2 ||S.
(7.17)
82
(7.18)
Let f (S, t) denote the value of the option which is replicated by the above portfolio, i.e., f = . Using Itos lemma,
f
f =
S +
S
f
2 2 f
+ S2 2
t
2
S
t.
(7.19)
f
S
!
f
2 2 2f
+ S
t.
t
2
S 2
(7.20)
(7.21)
where Le =
q
2
f
S.
S
(7.22)
2f
k
2
S||
Le S 2 2 t,
S
2
2
(7.23)
2f
f
2 2f
2
f
+ S 2 2 + Le S 2 2 + rS
rf = 0
S
t
2
S
2
S
(7.24)
2 2f
f
+ S 2 2 + rS
rf = 0,
t
2
S
S
(7.25)
2 = 2 [1 + Le sign()] , =
2
2f
S 2
(7.26)
H.E. Leland, Option pricing and replication with transaction costs, Journal of Finance,
83
7.4
Futures options
84
Futures options
Exchanges
Treasury-bond futures
CBOT
NYMEX
Copper futures
LME
CME
7.4.1
The time-dependence of a futures price is the same as that of the underlying asset.
Thus we suppose the futures price F follows the geometric Brownian motion:
dF
= F dt + F dZ
F
(7.27)
where F and F are, respectively, the constant expected rate of return and the
constant volatility, and dZ is the standard Wiener process. By Itos lemma, the
volatility of the futures price is given by
F =
1
F
S
= Ser =
F
S
F
(7.28)
85
where is the volatility for the underlying asset. From Itos lemma,
df =
f
f
1 2f 2 2
f
F F +
+
F dt +
F dZ
2
F
t
2 F
F
(7.29)
where f is the value of the derivative. For the case of a call option f = c and for
the case of a put option f = p.
Consider a portfolio composed of
1
f
+
F
derivative
number of futures contracts
(7.30)
at the opening of the contract because it costs nothing to enter into a futures
contract. Let d, df and dF be their changes of , f and F in time dt. In time
dt, the holder of the portfolio earns capital gains equal to df from the derivative
and income of
f
F dF
f
dF df
F
(7.31)
Substituting Eqs.(7.27) and (7.29) into Eq.(7.31) we find the change in wealth
!
f
1 2f 2 2
d =
F dt
t
2 F 2
(7.32)
which is non-stochastic and risk-free. Therefore the portfolio should give the
risk-free rate of return, r:
d = rdt.
(7.33)
We find the Black equation for futures derivatives by using Eqs.(7.30), (7.32) and
(7.33):
f
1 2f 2 2
+
F = rf.
t
2 F 2
This has the same form as Eq.(7.4) with q = r.
(7.34)
86
By solving the Black equation (7.34) for futures call and put options, we obtain
c = er [F N (d1 ) XN (d2 )]
p = er [XN (d2 ) F N (d1 )]
where
1
d1 =
F
2
ln
+
X
2
, d2 = d1 .
(7.35)
(7.36)
(7.37)