Lecture 9

BS Option Pricing
Model

1
 Continuous Time Option Pricing Models

• Assumptions of the Black-Scholes Option Pricing Model (BSOPM):

• No taxes
• No transactions costs
• Unrestricted short-selling of stock, with full use of short-sale proceeds
• Shares are infinitely divisible
• Constant riskless interest rate for borrowing/lending
• No dividends
• European options (or American calls on non-dividend paying stocks)
• Continuous trading
• The stock price evolves via a specific ‘process’ through time (more on this
later….)
Derivation of the BSOPM
• Specify a ‘process’ that the stock price will follow (i.e., all possible
“paths”)

• Construct a riskless portfolio of:

• Long Call
• Short D Shares
• (or, long D Shares and write one call, or long 1 share and write 1/D calls)

• As delta changes (because time passes and/or S changes), one must

maintain this risk-free portfolio over time

• This is accomplished by purchasing or selling the appropriate number

of shares.
 The BSOPM Formula
C  SNd1   KerTNd2 
where N(di) = the cumulative standard normal distribution function, evaluated at di,
and:

ln(S/K)  (r  σ 2 /2)T
d1 
σ T

d2  d1  σ T

N(-di) = 1-N(di)
 The Standard Normal Curve
• The standard normal curve is a member of the family of normal curves with μ =
0.0 and  = 1.0. The X-axis on a standard normal curve is often relabeled and
called ‘Z’ scores.

• The area under the curve equals 1.0.

• The cumulative probability measures the area to the left of a value of Z. E.g.,
N(0) = Prob(Z) < 0.0 = 0.50, because the normal distribution is symmetric.
 The Standard Normal Curve
The area between Z-scores of -1.00 and +1.00 is 0.68 or 68%.

Note also that the area What is N(1)?

to the left of Z = -1 What is N(-1)?
equals the area to the
right of Z = 1. This holds
for any Z.

The area between Z-scores of -1.96 and +1.96 is 0.95 or 95%.

What is the area in

each tail? What is
N(1.96)? What is N(-
1.96)?
 Cumulative Normal Table
Cumulative Probability for the Standard Normal Distribution
2nd digit of Z
z 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

-1.3 0.0968 0.0951 0.0934 0.0918 0.0901 0.0885 0.0869 0.0853 0.0838 0.0823

1.3 0.9032 0.9049 0.9066 0.9082 0.9099 0.9115 0.9131 0.9147 0.9162 0.9177

N(-d) = 1-N(d) N(-1.34) = 1-N(1.34)

Example Calculation of the BSOPM
Value
• S = $92
• K = $95
• T = 50 days (50/365 year = 0.137 year)
• r = 7% (per annum)
•  = 35% (per annum)

What is the value of the call?

Solving for the Call Price, I.
• Calculate the PV of the Strike Price:

Ke-rT = 95e(-0.07)(50/365) = (95)(0.9905) = $94.093.

• Calculate d1 and d2:

ln(S/K)  (r  σ 2 /2)T ln(92/95)  (0.07  0.1225/2)0 .137

d1  
σ T 0.35 0.137

ln(0.96842 )  0.01798  0.03209  0.01798

    0.1089
(0.35)(0.3 701) 0.12955

d2 = d1 – σT.5 = -0.1089 – (0.35)(0.137).5 = -0.2385

Solving for N(d1) and N(d2)
• Choices:
• Standard Normal Probability Tables (pp. 560-561)

• Excel Function NORMSDIST

• N(-0.1089) = 0.4566; N(-0.2385) = 0.4058.

• There are several approximations; e.g. the one in fn 8 (pg.

549) is accurate to 0.01 if 0 < d < 2.20:

N(d) ~ 0.5 + (d)(4.4-d)/10

N(0.1089) ~ 0.5 + (0.1089)(4.4-0.1089)/10 = 0.5467
Thus, N(-0.1089) ~ 0.4533
Solving for the Call Value
C = S N(d1) – Ke-rT N(d2)

= (92)(0.4566) – (94.0934)(0.4058)

= $3.8307.

Applying Put-Call Parity, the put price is:

P = C – S + Ke-rT = 3.8307 – 92 + 94.0934 = $5.92.

 Lognormal Distribution
• The BSOPM assumes that the stock price follows a “Geometric Brownian
Motion” (see
http://www.stat.umn.edu/~charlie/Stoch/brown.html
for a depiction of Brownian Motion).
• In turn, this implies that the distribution of the returns of the stock, at any
future date, will be “lognormally” distributed.
• Lognormal returns are realistic for two reasons:
• if returns are lognormally distributed, then the lowest possible return in any
period is -100%.
• lognormal returns distributions are "positively skewed," that is, skewed to the
right.

• Thus, a realistic depiction of a stock's returns distribution would have a

minimum return of -100% and a maximum return well beyond 100%. This is
particularly important if T is long.
The Lognormal Distribution

E ( ST )  S0 e T
2 2 T  2T
var ( ST )  S0 e (e  1)
BSOPM Properties and Questions
• What happens to the Black-Scholes call price when the call gets deep-
deep-deep in the money?

• How about the corresponding put price in the case above? [Can you
verify this using put-call parity?]

• Suppose  gets very, very close to zero. What happens to the call
price? What happens to the put price?
Hint: Suppose the stock price will not change from time 0 to time T. How
much are you willing to pay for an out of the money option? An in the
money option?
Volatility
• Volatility is the key to pricing options.

• Believing that an option is undervalued is tantamount to believing

that the volatility of the rate of return on the stock will be less than
what the market believes.

• The volatility is the standard deviation of the continuously

compounded rate of return of the stock, per year.
Estimating Volatility from
Historical Data
1. Take observations S0, S1, . . . , Sn at intervals of t (fractional years);
e.g., t = 1/52 if we are dealing with weeks; t = 1/12 if we are dealing
with months.
2. Define the continuously compounded return as:
 Si  ; on an ex-div day:  S  div 
ri  ln
S   ri  ln i 
 i1   Si1 
3. Calculate the average rate of return
4. Calculate the standard deviation,  , of the ri’s
5. Annualize the computed  (see next slide).
Annualizing Volatility
• Volatility is usually much greater when the market is open (i.e. the asset is
trading) than when it is closed.
• For this reason, when valuing options, time is usually measured in “trading
days” not calendar days.
• The “general convention” is to use 252 trading days per year. What is most
important is to be consistent.

σ annual  σ per trading day  252

σ annual  σ weekly  52

σ annual  σ monthly  12

Note that σyr > σmo > σweek > σday

Implied Volatility
• The implied volatility of an option is the volatility for which the BSOPM
value equals the market price.
• The is a one-to-one correspondence between prices and implied
volatilities.
• Traders and brokers often quote implied volatilities rather than dollar
prices.
• Note that the volatility is assumed to be the same across strikes, but it
often is not. In practice, there is a “volatility smile”, where implied
volatility is often “u-shaped” when plotted as a function of the strike
price.
Using Observed Call (or Put) Option Prices
to Estimate Implied Volatility
• Take the observed option price as given.
• Plug C, S, K, r, T into the BSOPM (or other model).
• Solving  this is the tricky part) requires either an iterative
technique, or using one of several approximations.

• The Iterative Way:

• Plug S, K, r, T, and ˆ into the BSOPM.
• Calculate ˆ (theoretical value)
• Compare ĉ (theoretical) to C (the actual call price).
• If they are really close, stop. If not, change the value of
and start over again.
Volatility Smiles
In this table, the option premium column is the average bid-ask price for
June 2000 S&P 500 Index call and put options at the close of trading on
May 16, 2000. The May 16, 2000 closing S&P 500 Index level was
1466.04, a riskless interest rate of 5.75%, an estimated dividend yield of
1.5%, and T = 0.08493 year.
call price put price
Strike Call price Call IV with  = 0.22 Put Price Put IV with  = 0.22
1225 1.5625 0.336 0.054
1250 2.34375 0.3283 0.153
1275 2.625 0.3021 0.39
1300 3.8125 0.292 0.904
1325 5.75 0.2855 1.919
1350 129.5 0.2834 124.335 8.25 0.2762 3.753
1375 107.625 0.2689 102.51 11.375 0.2641 6.809
1400 86.625 0.2534 82.353 14.875 0.2463 11.533
1425 67.625 0.2423 64.288 20.0625 0.2315 18.35
1450 50.6875 0.2324 48.648 29 0.2285 27.592
1475 35.625 0.2201 35.612 38.625 0.2152 39.437
1500 22.5 0.2036 25.178 50.875 0.2016 53.885
1525 14 0.1982 17.173 66.75 0.1923 70.762
1550 7.875 0.1912 11.291 85.375 0.1826 89.761
1575 4.375 0.1896 7.154 106.875 0.1788 110.505
1600 2.28125 0.1881 4.367 129.5 0.1668 132.6
1625 1.21875 0.1897 2.569 153.375 0.1512 155.684
1650 0.625 0.1912 1.457
Graphing Implied Volatility versus Strike (S = 1466):

Volatility Smiles

0.4
Implied Volatility

0.3
Call IV
0.2
Put IV
0.1
0
1200 1400 1600
Strike Price

2
S S
IV  α  β1ln    β 2 ln    ε
X X

Using the call data:

α= 2.17; β1 = -4.47; β2 = 2.52; R2 = 0.988
Dividends
• European options on dividend-paying stocks are valued by substituting the stock
price less the present value of dividends into Black-Scholes.
* -rT
C = S 0 N(d1 ) - Ke N(d2 )

Where: S* = S – PV(divs) between today and T.

• Also use S* when computing d1 and d2.

• Only dividends with ex-dividend dates during life of option should be included.
• The “dividend” should be the expected reduction in the stock price expected.
• For continuous dividends, use S* = Se-dT, where d is the annual constant dividend
yield, and T is in years.
 The BOPM and the BSOPM, I.
• Note the analogous structures of the BOPM and the
BSOPM:
C = SΔ + B (0< Δ<1; B < 0)

C = SN(d1) – Ke-rTN(d2)

• Δ = N(d1)
• B = -Ke-rTN(d2)
The BOPM and the BSOPM, II.
• The BOPM actually becomes the BSOPM as the number of periods
approaches ∞, and the length of each period approaches 0.
• In addition, there is a relationship between u and d, and σ, so that the
stock will follow a Geometric Brownian Motion. If you carve T years
into n periods, then:

σ T/n
u=e −1
−σ T/n
d=e −1
and, the probability of an uptick, p, must equal
μ T
p = 0.5 + 0.5
σ n
The BOPM and the BSOPM: An Example

• T = 7 months = 0.5833, and n = 7.

• Let μ = 14% and σ = 0.40
• Then choosing the following u, d, and p will make the stock follow the “right” process for
making the BOPM and BSOPM consistent:

u = e0.40 0.0833 − 1 = 0.1224

d = e−0.40 0.833 − 1 = −0.10905
and, the probability of an uptick, p, must
equal
0.14
p = 0.5 + 0.5 0.0833 = 0.5505
0.40
 Generalizing the BSOPM
• It is important to understand that many option pricing models are related.

• The BSOPM itself, however, can be generalized to encapsulate several

important pricing models.

• All that is needed is to change things a bit by adding a new term, denoted by
b.
 The Generalized Model Can Price
European Options on:
1. Non-dividend paying stocks [Black and Scholes (1973)]

2. Options on stocks (or stock indices) that pay a continuous dividend

[Merton (1973)]

3. Currency options [Garman and Kohlhagen (1983)]

4. Options on futures [Black (1976)]

The Generalized Formulae are:

Cgen = Se N  d1  - Ke -rTN  d2 

b-r  T

Pgen = Ke-rTN  -d2  - Se N  -d1 

b-r  T

ln(S/K)  (b   2 /2)T
d1 
 T

ln(S/K)  (b   2 /2)T
d2   d1   T
 T
By Altering the ‘b’ term in These
Equations, Four Option-Pricing
Models Emerge.
By Setting: Yields this European Option Pricing Model:

b=r Stock option model, i.e., the BSOPM

b=r-q Stock option model with continuous

dividend yield, q

b = r – rf Currency option model where rf is the

foreign risk-free rate
American Options

• It is computationally difficult to value an American

option (a call on a dividend paying stock, or any
put).
• Methods:
• BOPM
• Numerical methods
• Approximations
• Of course, computer programs are most often
used.
 Black’s Approximation for Dealing with
Dividends in American Call Options
Set the American price equal to the
maximum of two European prices:
1. The 1st European price is for an option
maturing at the same time as the American
option
2. The 2nd European price is for an option
maturing just before the final ex-dividend
date