7. The Black-Scholes-Merton Model
7.1 Assumption
7.1.1 Stock Price Movements
The stock prices are often modeled by geometric Brownian motion (GBM).
The stock price S T S_T ST, is log-normally distributed. A variable that is log-normally distributed has a minimum of zero and conforms to prices better than the normal distribution which would produce negative prices.
If the stock price is log-normally distributed, the continuously compounded annual return of the stock is normally distributed.
7.1.2 Historical Volatility
Square root rule: the volatility for short periods of time can be scaled to longer periods in time by the square root rule.
σ
J
−
p
e
r
i
o
d
s
=
σ
1
−
p
e
r
i
o
d
×
J
0.5
\sigma_{J-periods}=\sigma_{1-period}\times J^{0.5}
σJ−periods=σ1−period×J0.5
J J J number of the trading periods in a year.
7.1.3 Assumption
The option should be European-style.
There are no dividends during the life of the option.
The underlying stock price follows lognormal distribution and the rate of return of the stock follows normal distribution.
The continuously compounded risk-free rate is constant and known.
The volatility of the stock return is constant and known.
The market is frictionless.
- The short selling is permitted and continuous trading is available. The underlying asset is divisible.
- There are no transaction costs, no taxes and no regulatory constraints.
- There is no-arbitrage opportunities in the market.
7.2 Pricing European Option
7.2.1 European Option on A Non-dividend Stock
Call
=
S
0
N
(
d
1
)
−
X
e
−
r
T
N
(
d
2
)
\text{Call}=S_0N(d_1)-Xe^{-rT}N(d_2)
Call=S0N(d1)−Xe−rTN(d2)
Put
=
−
S
0
N
(
−
d
1
)
+
X
e
−
r
T
N
(
−
d
2
)
\text{Put}=-S_0N(-d_1)+Xe^{-rT}N(-d_2)
Put=−S0N(−d1)+Xe−rTN(−d2)
d
1
=
I
n
(
S
0
X
)
+
(
r
f
+
σ
2
2
)
T
σ
T
d_1=\frac{In(\frac{S_0}{X})+(r_f+\frac{\sigma^2}{2})T}{\sigma\sqrt{T}}
d1=σTIn(XS0)+(rf+2σ2)T
d
2
=
I
n
(
S
0
X
)
+
(
r
f
−
σ
2
2
)
T
σ
T
d_2=\frac{In(\frac{S_0}{X})+(r_f-\frac{\sigma^2}{2})T}{\sigma\sqrt{T}}
d2=σTIn(XS0)+(rf−2σ2)T
d
2
=
d
1
−
σ
T
d_2=d_1-\sigma\sqrt{T}
d2=d1−σT
- N ( X ) N(X) N(X): Standard normal cumulative distribution function.
- N ( d 1 ) N(d_1) N(d1): Delta of call option
- N ( d 2 ) N(d_2) N(d2): Risk-neutral probability of call exercise.
7.2.2 European Option on A Dividend-paying Stock
Discrete dividends
Call
=
(
S
0
−
PVD
)
N
(
d
1
)
−
X
e
−
r
T
N
(
d
2
)
\text{Call}=(S_0-\text{PVD})N(d_1)-Xe^{-rT}N(d_2)
Call=(S0−PVD)N(d1)−Xe−rTN(d2)
Put
=
−
(
S
0
−
PVD
)
N
(
−
d
1
)
+
X
e
−
r
T
N
(
−
d
2
)
\text{Put}=-(S_0-\text{PVD})N(-d_1)+Xe^{-rT}N(-d_2)
Put=−(S0−PVD)N(−d1)+Xe−rTN(−d2)
Continuously compound dividends:
Call
=
S
0
e
−
q
T
N
(
d
1
)
−
X
e
−
r
T
N
(
d
2
)
\text{Call}=S_0e^{-qT}N(d_1)-Xe^{-rT}N(d_2)
Call=S0e−qTN(d1)−Xe−rTN(d2)
Put
=
−
S
0
e
−
q
T
N
(
−
d
1
)
+
X
e
−
r
T
N
(
−
d
2
)
\text{Put}=-S_0e^{-qT}N(-d_1)+Xe^{-rT}N(-d_2)
Put=−S0e−qTN(−d1)+Xe−rTN(−d2)
Effects on options: Dividend payment will increase put values and decrease call values.
7.2.3 Implied Volatility
Volatility is the only parameter which we cannot get directly in the market. Four inputs in fives for BSM model is observable ( S S S, X X X, T T T, R R R), and if the option market price is available, the volatility can be calculated with the BSM model. The volatility of underlying asset return “implied” in the option market price.
Volatility estimation using historical data is not always representative of current market (backward-looking) while implied volatility is forward-looking.
810
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
