Chapter 13 Binomial Tree Complete Version Fall 2022-20221101

Introduction to
Binomial Trees
Chapter 12
Prof. Joseph K. W. Fung

FIN B488F Fall 2022
12.1
A brief history of financial theory
development
 Markowitz: mean-variance analysis for securities
portfolio 1950’s; ranking more than pricing (valid for
both risk-averse & risk-neutral investors GM)
 Sharpe: CAPM late 1960’s, valuation of risky asset
via a model for the required rate of return. V =
E(CF) / 1+RRR; CAPM=>RRR (RRR is proportional
to beta/market risk); assumes risk-averse investors
(separation theorem)
 Black-Scholes-Merton: option pricing model, 1973
(timely, CBOE Chicago board option exchange). V’ =
Ep (CF) / 1+Rf; applicable to all investors (via pure
arbitrage)
Prof. Joseph K. W. Fung 12.2
FIN B488F Fall 2022
A comparison between Sharpe’s
and BSM valuation frameworks
 Markowitz (risk-neutral & risk-averse): mean-variance criteria
provide ranking for assets
 Sharpe (requires risk-aversion):
V = E(CF) / 1+RRR; E(CFs) are based on the subjective and/or

objective belief of the future cashflows; as the CFs are risky or
uncertain, the discount rate RRR reflects the level of risk; E(.)
is based on true or natural probability
Black-Scholes-Merton (independent of risk preference):
V’ = Ep (CF) / 1+Rf; the discount rate is the risk-free rate of
interest (that is common for all asset); Ep(.) is an artificial
probability called risk-neutral probability measure. In
mathematical literature, the pricing model is called martingale
pricing model; Ep(.) is called a equivalent martingale

FIN B488F Fall 2022
What is the price of a European
call option?
 A stock price is currently $20 (S)
 A 3-month (OTM) call option on the stock has a strike price of $21 (X)
 The 3-month risk-free interest rate is 12% per annum (r).
 We know the price of the call is between its upper and lower bounds; i.e.,
(lower bound) Max[20-21e – 0.120.25 ,0] < C(X=21,S=20) < 20 = S (upper bound)
 0 < C < 20
 Boundary values are highly robust: but the band is wide. Need more information to fix the call price!

FIN B488F Fall 2022
A Simple Binomial Model
 A stock price is currently $20
 In three months it will be either $22 or $18
 Su = Su = $22, u = 1.1 (up-jump factor)
 Sd = Sd = $18, d = 0.9 (down-jump factor)

Up state stock
Price: Su = $22
Stock price: S = $20
Down state stock
Price: Sd = $18

FIN B488F Fall 2022
A Call Option (Figure 12.1, page 268)
A 3-month call option on the stock has a strike price of $21
Su = $22
Option payoff/value = $1
S = $20
Option Price=?
Sd = $18
Option payoff/value = $0

FIN B488F Fall 2022
Setting Up a Risk-free Portfolio
 Consider the Portfolio (sell 1 call and hedge it with a fraction of a
share; sort of buy-write / covered call):
 long  shares ( 0 <
 short 1 call option $22 – $1
$18
 Portfolio is “risk-free” when $22– $1 = $18 or  = 0.25

 Risk-free means that the values of the portfolio in both states (up state or down state) are equal
44 = 0.25
  exist and unique


FIN B488F Fall 2022
Valuing the Portfolio
(Risk-Free Rate is 12%)
 The risk-free portfolio is:
 Long 0.25 shares
 Short 1 call option
 The value of the portfolio in 3 months is
Upstate portfolio value = $22 0.25 – $1 = $4.50
 Down state value = $18 (0.25) = $4.50
 The present value (as discounted by the risk-free rate of
interest) of the portfolio today is $4.5e –
0.120.25
= $4.367
FIN B488F Fall 2022
Valuing the Option (no-arbitrage)
 The portfolio that long 0.25 shares and short
1 option is worth $4.367 today
 S – 1C = $4.367
 S = $20, S = $5 (0.25 x $20 )
 C* = S–$4.367 = $5 – $4.367 = $0.633
 “Fair” value of the option is $0.633
 C* is positively related to S and 
FIN B488F Fall 2022
Arbitrage with an Overpriced Call (c > c* no-
arbitrage value)
If c = $0.8 > c* = $0.633 (c is overpriced relative to S)
Call is high & stock is low since the fair value of the call is determined by the
price of the stock. Strategy: Short 1 Call and Buy 0.25 Shares (this creates a
sure future payoff of $4.50)
CFT
Transactions CF0 ST > X ST < X
(ST = $22) (ST = $18)
Borrow (PV $4.5) $4.367 at 12%
+$4.367 – $4.5 – $4.5
for 3m
Short 1 Call (X = $21) +$0.8 – $1 $0
Buy 0.25 Shares of Stock – $5 +$5.5 +$4.5
+$0.167 $0 $0
FIN B488F Fall 2022
Arbitrage with an Underpriced Call (c < c*)
If c = $0.5 < c* = $0.633
Call is low & stock is high. Strategy: Buy 1 Call and Short 0.25 Shares (this creates
a future liability of $4.50)
CFT
Transactions CF0 ST > X ST < X
(ST = $22) (ST = $18)
Short 0.25 Shares of Stock +$5 – $5.5 – $4.5

Long 1 Call (X = $21) – $0.5 +$1 $0
Lend $4.367 at 12% for 3m – $4.367 +$4.5 +$4.5
+$0.133 $0 $0
FIN B488F Fall 2022
What is a call under the model?
 Short 1 Call + 0.25 Stock create a sure cash
flow of $4.50 after 3-month
 - C + 0.25 S = PV($4.50) = bond (par=4.5)
 C = 0.25 S – Bond; delta dynamic trading
 Margin purchase again! (a “futures”?)
 Hence, the call is a leveraged stock position in
the one-period case; S = 4C + 4 Bond
 Note that under put-call-parity, a call is
composed of a bond, a put, and stock
FIN B488F Fall 2022
Generalization (Figure 12.2, page 269)
A derivative ( f ) lasts for time T and is

dependent on a stock (or any underlying
asset) S
Su
ƒu
S
ƒ Sd
ƒd

FIN B488F Fall 2022
Generalization (continued)
 Consider the portfolio that is (“long” if f is a call or “short” if f is
a put)  S and short 1 derivative
Su– ƒu
Sd– ƒd
 The portfolio is risk-free when Su– ƒu = Sd– ƒd or
ƒu  f d

Su  Sd
FIN B488F Fall 2022
 Value of the portfolio at time T is (a sure thing) Su–
ƒu = Sd– ƒd
 Since the portfolio is risk-free, value of the portfolio today is

(Su – ƒu )e–rT
 The value of the portfolio today is S– f = (Su –

ƒu )e–rT
 Hence, ƒ = S– (Su– ƒu )e–rT
 1st term RHS is the hedge with S, 2nd term is a fixed income
position (leverage stock position if f is a call option)

FIN B488F Fall 2022
What is finance?
 We sell hopes (call option that will end up out of
the money ）
 We profit from people’s fear （ Sell put to make
people feels safer but the put may end up out of
the money)
 Either way, the people buy options for nothing
 Who produce option? Who sells options?
Investment bankers

FIN B488F Fall 2022
A derivative can be
replicated/mimicked by a
combination of S and a “Bond”
 ƒ = S– (Su– ƒu )e–rT
The derivative f can be replicated by a (cash spot)
position in S and a short position in
(Su– ƒu )e–rT of fixed income security (I.E., a
loan or deposit of that amount

FIN B488F Fall 2022
 ƒ = S– (Su– ƒu )e–rT
 RHS: 1st term is a stock position and 2nd term

is a bond position; hence a (long) derivative
position is a leverage stock (asset) position
 For Call, > 0; call is a leveraged long stock
position
 For Put, < 0; put is a leveraged short stock
position

FIN B488F Fall 2022
 Substituting for  we obtain (verify it!)
ƒ = Ep (f) e–rT = [p ƒu + (1 – p )ƒd ]e–rT
e d
rT
where p
ud
ue rT
1 p 
ud
FIN B488F Fall 2022
Conditions for a probability measure
(1): non-negativity 0< p<1
(2): sum of the probabilities of all disjoint events = 1
(3): completeness
Assumption: u > erT > d (the following proof is by way of contraction)
 If erT > u
 Risk-free return > maximum stock return
 nobody speculates in the stick market, hence erT < u
 If d > erT
 return in stock market in the worst case > risk-free return
 nobody invests in the fixed income market, hence d < erT

FIN B488F Fall 2022
Risk-Neutral Valuation
 ƒ = [ p ƒu + (1 – p )ƒd ]e-rT
 The variables p and (1– p) can be interpreted as the risk-
neutral probabilities of up and down movements
 The value of a derivative is its expected payoff in a risk-
neutral world discounted at the risk-free rate
Su
p
ƒu
S
ƒ (1 Sd
–
p) ƒd
FIN B488F Fall 2022
Risk Neutral Probability p
Investors are risk neutral if
Se rT
 E ST   pSu  1  p Sd
e rT  d u  e rT
pSu  1  p Sd  Su  Sd  Se rT
ud ud
Hence p & 1 – p are called risk-neutral probabilities

Does p equal to the “real” chance that the market goes up?
FIN B488F Fall 2022
Irrelevance of Stock’s (and
derivative’s) Expected Return
(e.g., CAPM required ret) that is
where Samuelson got lost
When we are valuing an option in
terms of the underlying stock the
expected returns on the stock or on the
option are irrelevant
Risk preference irrelevant ！ The only
assumption is investors/traders are
GREEDY!! (prefer more to less to
motivate pure arbitrage)
FIN B488F Fall 2022
Original Example Revisited
Su = $22
p ƒu = 1
S = $20
ƒ
(1– Sd = $18
p) ƒd = 0
 Since p is a risk-neutral probability: definition
 $20e0.12 0.25 = $22p + $18(1 – p ); p = 0.6523
 Alternatively, we can use the formula
e rT  d e 0.120.25  0.9
p   0.6523
ud 1.1  0.9
FIN B488F Fall 2022
Valuing the Option with risk-neutral
probabilities (“equivalent martingale”)
Su = $22
523 ƒu = 1
0. 6
S = $20
ƒ
0.34 Sd = $18
77
ƒd = 0
The value of the option is
= e–0.120.25 (0.65231 + 0.34770) = $0.633
Note that the pricing formula fails to show how to trade

FIN B488F Fall 2022
Δ as the hedge ratio
 S0= 20, Su= 22, Sd = 18, Cu= 1, r = 12%, T = 1/4
$22Δ – $1 = $18Δ
4Δ = 1
Δ = 0.25  (long 0.25 shares per 1 short call)
 If stock price goes up, value of portfolio = $22(0.25) – $1 = $4.5
 If stock price drops, value of portfolio = $18(0.25) = $4.5
 Present value of portfolio = $4.5e-0.12/4 = $4.367

FIN B488F Fall 2022
Δ as the hedge ratio
SuΔ – fu = SdΔ – fd
Δ = (fu – fd) / (Su – Sd)
Δ: ratio of change in option (any other derivative)

price to the change in stock price.
Delta shows how the value of a derivative is affected

by the underlying asset; hence the derivative can be
priced relative to S
FIN B488F Fall 2022
A Two-Step Example: Call (X=21)
Figure 12.3, page 274
Suu = $24.2
Su = $22
S = $20 Sud = $19.8
Sd = $18
Sdd = $16.2
 Each time step is 3 months; total length 6 m.

 u = 1.1; d = 0.9; r = 12%; T = 0.25; p = 0.6523
 Path independence: Sud = Sdu
FIN B488F Fall 2022
Valuing a Call Option (X=21)
Su = $22 D Suu = $24.2
fu = $2.0257 fuu = $3.2
S = $20 B E Sud = $19.8
f = $1.2823 A
C fud = $0
F
Sd = $18 Sdd = $16.2
fd = $0 fdd = $0
 Value at node B = e–0.120.25(0.65233.2 + 0.34770) =

$2.0257
 Value at node C = e–0.120.25(0.65230 + 0.34770) = $0
 Value at node A = e–0.120.25(0.65232.0257 + 0.34770) =
$1.2823

FIN B488F Fall 2022
Two-step binominal option price:
a generalization
fu p fuu
p
f 1–p fud
p
1–p
fd 1–p fdd
fu = [ pfuu + (1 – p)fud ] e-rT
fd = [ pfud + (1 – p)fdd ] e-rT
f = [ pfu + (1 – p)fd ] e-rT
f = e-r2T [ p2fuu + 2p(1 – p)fud + (1 – p)2fdd ]

FIN B488F Fall 2022
Two-period binomial option
pricing

f  e  2 rt p 2 f uu  2 p (1  p ) f ud  (1  p ) 2 f dd 
e rT  d
p
ud

FIN B488F Fall 2022
n-period model
Assume that each period is of length , n = T
n 
C  r n n! j n j j n j

p 1  p  Max u d S  X ,0  
 j 0 j!n  j ! 
where r = 1+ risk free interest rate per period
n!
p j 1  p 
n j
j!n  j ! is the (binomial) probability that the
value of the call is equal to

Max u j n j

d S  X ,0 after n 
periods or (n moves) when the option expires

FIN B488F Fall 2022
Nature of options’ Truncated Risk
Define “a” as the minimum no. of up moves such that
u a d na S  X and the call finishes in-the-money,

then
 n 
C  r 
n n! j n j

p 1  p  u d S  X 
j n j

 i  a j!n  j ! 
 n n! n j  u d
j n j
  n n! n j 
 S  p 1  p  
j
  Xr 
n
p j 1  p  
 j  a j!n  j !  j  a j!n  j !
n
 r  

FIN B488F Fall 2022
n-period model
na
au d a
SX
Take natural logarithm of both sides, after re-arranging, we have,
a ln u  n  a ln d  S  X
aln u  ln d   X  S  n ln d
a is the smallest non-negative integer greater than


ln X / Sd n 
ln u / d 
wherea n

FIN B488F Fall 2022
n n!  u j n j
d    n
n! n j 
C  S  p 1  p   n   Xr  p 1  p  
j n  j n j
 j  a j!n  j !  r   j  a j!n  j ! 
The first bracketed term is the number of shares or the hedge ratio ( Δ)
ƒ = S– (Su– ƒu )e–rT
The second term is the size of the loan in the replicating portfolio
The bracketed expression in the second term is the total prob. That the
option will end up in the money and will be exercised (i.e., the exercise
probability)

FIN B488F Fall 2022
 n n! n j  u d
j n j
  n n! n j 
C  S  p 1  p  
j
  Xr 
n
p 1  p  
j
 j  a j! n  j !  r n
  j  a j!n  j ! 
The first bracketed term can be simplified as
n j
 pu   1  p d 
n j
n!
j a
  
j!n  j !  r   r


u d
Let p '  p; 1  p '  (1  p ); and 0  p '  1
r r
 n n! n j   n n! n j 
C  S  p 1  p   Xr 
j n
p 1  p  
j
 j  a j!n  j !   j  a j!n  j ! 

FIN B488F Fall 2022
 n n! n j   n n! n j 
C  S  p 1  p   Xr 
j n
p 1  p  
j
 j  a j!n  j !   j  a j!n  j ! 
u
p '  p; 0  p '  1
r
(r  d ) r
and p  
(u  d ) u
u
Hence p is always positive and less than 1.
r

FIN B488F Fall 2022
C  Sa; n; p  Xr a; n; p 
' n

 a; n; p '
 is the complimentary probability or delta
a; n; p  is the probability of exercise

a is the smallest non-negative integer greater than ln X /Sd n

ln u / d 
If a>n, then C=0
If S is large relative to X, a is small and the probability of

exercise is high.
p  (r  d ) /(u  d ) and p  (u / r ) p
FIN B488F Fall 2022
Delta
 Delta () is the ratio of the change in the price of a

stock option to the change in the price of the
underlying stock
 The value of  varies from node to node
ƒu  f d

Su  Sd

FIN B488F Fall 2022
“Deltas” in a 2-period binomial model
Su = $22 D Suu = $24.2
fu = $2.0257 fuu = $3.2
S0 = $20 B E Sud = $19.8
A
f = $1.2823 C fud = $0
F
Sd = $18 Sdd = $16.2
fd = $0 fdd = $0
fu – fd (2.0257 – 0)
 at node A () = = = 0.5064
Su – Sd (22 – 18)
fuu – fud (3.2 – 0)
 At node B (u) = = = 0.7273
Suu – Sud (24.2 – 19.8)
fud – fdd (0 – 0)
 at node C (d) = = =0
Sud – Sdd (19.8 – 16.2)
FIN B488F Fall 2022
Dynamic hedging: why?
 At A: delta = 0.5064; c=1.2823
 S = 20
 At B: S = 22 and c = 2.0257
 At A: (riskfree) portfolio = 0.5064 (20) – c
 At B: change in stock value = 0.5064 ($2) = 1.0128; change in
c value = 2.0257-1.2823 = 0.7434
 At D: 0.5057 (4.2) = 2.1219; change in call = 3.20-1.2823 =
1.9177 (lose money!)
 Reason: delta increases for call when S rises
 CONVEXITY (GAMMA RISK) = delta is a function of S
 Need to change hedge ratio: otherwise, under-hedge if S rises
and over-hedge if S drops
 Hedging a call requires an adjustment in delta when S changes
FIN B488F Fall 2022
Arbitrage in a 2-step binomial model
If c = 1.5 > c* = 1.2823, the call is overpriced
Strategy: Short 1 Call and buy 0.5064) Shares
Transactions CF0
Short 1 Call +$1.5
-$10.128
Buy  Shares
(- 0.5064 x $20)
Borrow $8.8457 at 12% for 6 months +8.8457

+0.2177

FIN B488F Fall 2022
Portfolio rebalancing at node B
At node B, u= 0.7273, to reach the new hedge ratio, need to

purchase an additional 0.2209 (u – = 0.7273 – 0.5064)
shares. Finance the purchase with additional borrowing
Transactions at node B (Su = S3 = 22)
-$4.8598
Buy (u –  Shares
-(0.2209 x $22)
Borrow $4.8598 at 12% for 3 months +$4.8598

FIN B488F Fall 2022
Payoff from the overall portfolio at
option expiration (X=$21)
CF6
(Suu = S6 (Sud = S6
= $24.2) = $19.8)
Short 1 Call -$3.2 $0
u Shares: value=0.7273 x S6 +$17.600 +$14.400
Repayment of the initial (6-month) loan -$9.392 -$9.392

Repayment of the second (3-month) loan -$5.008 -$5.008
$0 $0
FIN B488F Fall 2022
Portfolio rebalancing at node C
At node C, d= 0, to reach the new hedge ratio, need

to sell all the 0.5064 () shares.
Transactions at node C (Sd = S3 = $18)
+$9.115
Sell  Shares
(0.5064 x $18)
Lend $9.115 at 12% for 3 month -$9.115

FIN B488F Fall 2022
option expiration
CF6
(Sud = S6 (Sdd = S6
= $19.8) = $16.2)
Short 1 Call 0 0
Repayment of the initial (6-month) loan -$9.392 -$9.392
The second (3-month) investment +$9.392 +$9.392
0 0

FIN B488F Fall 2022
Summary on dynamic portfolio
rebalancing
 If S up, delta or hedge ratio up, need to buy more
shares to maintain the hedge (delta neutral)
 If S down, delta or hedge ratio down, need to
sell/unload some or all shares to maintain the hedge
(delta neutral)
An alternative strategy
 If S up, need to buy more call to maintain the delta
neutral position (buy call when call price goes up)
 If S down, need to sell more calls to maintain the
delta neutral position (sell calls when call price drops)
 These may show the cost for selling options for
investment banks
FIN B488F Fall 2022
dynamic portfolio rebalancing (example)
 Initial call delta = 0.5
 Sell 1000 calls, buy 500 shares
 Now S rises, new delta = 0.75
 To adjust the hedge according to the higher delta:
1. Buy more stock: required number of share to hedge the

call = 750 shares, need to buy 250 shares
2. buy back some calls, with 500 shares and new call delta
of 0.75, can hedge only 500/0.75 = 667, buy back 333
calls, 500/667 = 0.75
Note that either way, the option seller has to buy more stock
or option when they are the rise
Similarly, if S drops, delta drops, IB sell some stock or sell
more call when they are falling
FIN B488F Fall 2022
Arbitrage in a 2-step binomial model
If c = 1 < c* = 1.2823, the call is underpriced
Strategy: Long 1 Call and Short 0.5064) Shares
Transactions CF0
Long 1 Call -$1
+$10.128
Short  Shares
(0.5064 x $20)
Lend $8.8457 at 12% for 6 months -8.8457

+0.2823

FIN B488F Fall 2022
Portfolio rebalancing at node B
At node B, u= 0.7273, to reach the new hedge ratio, need to
short an additional 0.2209 (u – = 0.7273 – 0.5064) shares.
Invest the additional cash inflow
Transactions at node B (Su = S3 = 22)
$4.8598
Short (u –  Shares
(0.2209 x $22)
Lend $4.8598 at 12% for 3 months -$4.8598

FIN B488F Fall 2022
option expiration
CF6
(Suu = S6 = (Sud = S6 =
$24.2) $19.8)
Long 1 Call +$3.2 $0
u Shares -$17.600 -$14.400
The initial (6-month) investment +$9.392 +$9.392
The second (3-month) investment +$5.008 +$5.008
$0 $0
FIN B488F Fall 2022
Portfolio rebalancing at node C
At node C, d= 0, to reach the new hedge ratio, need to

buy back the 0.5064 () shares. Finance the purchase
with additional borrowing
Transactions at node C (Sd = S3 = $18)
-$9.115
Buy  Shares
(- 0.5064 x $18)
Borrow $9.115 at 12% for 3 month +$9.115

FIN B488F Fall 2022
option expiration
CF6
(Sud = S6 (Sdd = S6
= $19.8) = $16.2)
Long 1 Call 0 0
The initial (6-month) investment +$9.392 +$9.392
Repayment of the second (3-month) loan -$9.392 -$9.392
0 0

FIN B488F Fall 2022
Investment banks and market momentum
(world federation exchanges vol/turnover)
 IB short call and hedge with long stock
 Market goes up, buy stock because delta

rises
 Market drops, sell stock because delta
decreases
 In both cases, the investment bank add
momentum in both up and down market
 Testable hypothesis: markets with active
options trading have higher volatility
FIN B488F Fall 2022
market momentum
 Momentum  positive return correlation In
both cases
 If derivative market (due to manipulation
and/or the technical requirement for
dynamic portfolio rebalance) adds to
market momentum, then markets with
active and large derivative trading would
have greater price momentum; that means
that the returns have higher auto-
correlation; test that with trading rules
FIN B488F Fall 2022
IB and market momentum
 Short put and hedge with short stock
(futures) position
 Market goes up, reduce the short position,
buy stock
 Market drops, increase the short position,
sell stock
 In both cases, they add momentum to the
market
FIN B488F Fall 2022
A Put Option Example (X=52)
Suu = $72
Su = $60
S = $50 Sud = $48
Sd = $40
Sdd = $32
 Each time step is 1 year

 u = 1.2; d = 0.8; r = 5%; T = 1; p = 0.6282

FIN B488F Fall 2022
Valuing Put Option (X=52)
Su = $60 D Suu = $72
fu = $1.4147 fuu = $0
S = $50 B E Sud = $48
f = $4.1923 A
C fud = $4
F
Sd = $40 Sdd = $32
fd = $9.4636 fdd = $20
 Value at node B = e–
0.120.25
(0.62820 + 0.37184) = $1.4147
 Value at node C = e–
0.120.25
(0.62824 + 0.371820) = $9.4636
 Value at node A = e–
0.120.25
(0.62821.4147 + 0.37189.4636) = $4.1923
FIN B488F Fall 2022
What Happens if the Put Option is American
(Figure 12.8, page 278)
Su = $60 D Suu = $72

fu = $1.4147 fuu = $0
S = $50 B E Sud = $48
f = $5.0894 A
C fud = $4
F
Sd = $40 Sdd = $32
fd = $12 (ex.) fdd = $20
 Value at node B = e–0.120.25(0.62820 + 0.37184) = $1.4147

 Value at node C = $12 (early exercise)
 Value at node A = e–0.120.25(0.62821.4147 + 0.371812) = $5.0894
 American option ($5.0894) is more valuable than European option
($4.1923 ) if there is a chance of early exercise
 Difference between the two prices is the value of early exercise option
FIN B488F Fall 2022
Value of the put option at node “C”
 Value of the option without early exercise:
a = [ pSud + (1 – p) Sdd ]e–rT
= (0.62824 + 0.371820)e–0.120.25 = $9.4636

 Value of the option with early exercise:
b = Max [X – Sd , 0] = $12
 Value of the option at note c
= Max [a, b] = $12

FIN B488F Fall 2022
Delta
Example: Valuing a Put Option (Figure 11.7, page 252)
 The  over the first time step:
= (1.4147 – 9.4636) / (60 – 40) = – 0.4024

 The  over the second time step:
 If there is an upward movement over the first step:
= (0 – 4) / (72 – 48) = – 0.1667

 If there is an downward movement over the first step:
= (4 – 20) / (48 – 32) = – 1.0000

FIN B488F Fall 2022
The Probability of an Up Move
a–d
p =
u–d
a = e r t  for a non-dividend paying stock
a = e(r - q)t  for a stock index where q is the

dividend yield on the index
a = e(r – rf)t  for a currency where rf is the foreign

risk-free rate
a = 1  for a futures contract (exp futures
return is zero)
FIN B488F Fall 2022
Choosing u and d
One way of matching the volatility is to set
u  e t
d  1 u  e  t
where  is the volatility andt is the length

of the time step. This is the approach used
by Cox, Ross, and Rubinstein
Prof. Joseph
Options, K. W.and
Futures, Fung
Other Derivatives, 8th Edition, Copyright © John C. Hull 2012 63
FIN B488F Fall 2022
Size up u and d
One way of matching the volatility is to set
u  e t
d  1 u  e  t
example:  = 20%
t = 1 u  e 0.2 1
 1.22
d  1 1.22  0.82
Prof. Joseph
Options, K. W.and
Futures, Fung
FIN B488F Fall 2022
Size up u and d
t u  e 0.2 1/1386
1: u =  1.005
d  1 / 1.005  0.995
Prof. Joseph
Options, K. W.and
Futures, Fung
FIN B488F Fall 2022
Girsanov’s Theorem
 Volatility is the same in the real world and
the risk-neutral world
 We can therefore measure volatility in the
real world and use it to build a tree for an
asset in the risk-neutral world
Prof. Joseph
Options, K. W.and
Futures, Fung
FIN B488F Fall 2022
Assets Other than Non-Dividend
Paying Stocks
 For options on stock indices, currencies

and futures the basic procedure for
constructing the tree is the same except
for the calculation of p
Prof. Joseph
Options, K. W.and
Futures, Fung
FIN B488F Fall 2022
Proving Black-Scholes-Merton from
Binomial Trees (Appendix to Chapter 12)
n
n!
ce  rT

j  0 ( n  j )! j!
p j (1  p ) n  j max(S 0u j d n  j  K , 0)
Option is in the money when j >  where

n ln( S 0 K )
 
2 2 T n
so that
c  e  rT ( S 0U1  KU 2 )
where
n!
U1   p j (1  p ) n  j u j d n  j
j   ( n  j )! j!
n!
U2   p j (1  p ) n  j
j   ( n  j )! j!
Prof. Joseph
Options, K. W.and
Futures, Fung
FIN B488F Fall 2022
Proving Black-Scholes-Merton
from Binomial Trees continued
 The expression for U1 can be written

U1  [ pu  (1  p )d ]n 
n!
  j
p* 1  p* 
n j
 e rT 
n!
  j
p* 1  p* 
n j
j   ( n  j )! j! j   ( n  j )! j!
pu
p* 
where pu  (1  p )d
 Both U1 and U2 can now be evaluated in terms of the

cumulative binomial distribution
 We now let the number of time steps tend to infinity
and use the result that a binomial distribution tends to
a normal distribution by CENTRAL LIMIT THEOREM
Prof. Joseph
Options, K. W.and
Futures, Fung
FIN B488F Fall 2022
Impacts of changes in delta on risk
management (1)
 The hedge of an option position has to be
adjusted (rebalance) dynamically
according to asset price movements
 Sellers or buyers: call seller adjust the
long asset position, call buyer adjust the
short asset position; similarly put seller
adjust the short asset position and put
buyer adjust the long asset position

FIN B488F Fall 2022
Impacts of changes in delta on risk
management (2)
 Sellers of options will adjust the hedge in such a
way that will add to market momentum
 Both call and put seller need to increase (decrease)
size (+/-) asset holding to maintain delta neutrality
when asset price rises (drops); these imply that the
options sellers are buying when market rises and
selling when market drops; hence will increase the
market impact cost in the dynamic hedge (market is
not a perfect market where the trader can buy and
sell an infinite amount of an asset at the same price;
note that the supply curve is horizontal in a perfect
market)
FIN B488F Fall 2022
More on hedging option position (3)
 What if there is a big jump (gamma risk which
is associated with change in delta) in the asset
price over-night?
 For example: S&P was closed at 1850 but
open at 1900 with a 50 points jump
 Buy S&P at 1900?
 What about the price of the calls being sold
yesterday?
 If the calls were sold for 37 points with a delta
of around 0.5, the price should have gone up
by 0.6 (50) = 30 points
FIN B488F Fall 2022
Ch 12: exercises ???
 1, 2, 4, 5,
 9, 10, 11, 12, 16, 17

FIN B488F Fall 2022
 Financial analyst journal
 Journal of portfolio management
 Derivatives and hedge fund
 Journal of futures markets
 Review of futures markets
 Journal of derivatives

FIN B488F Fall 2022

Chapter 13 Binomial Tree Complete Version Fall 2022-20221101

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 13 Binomial Tree Complete Version Fall 2022-20221101

Uploaded by

Copyright:

Available Formats

Introduction to

Prof. Joseph K. W. Fung

V = E(CF) / 1+RRR; E(CFs) are based on the subjective and/or

Prof. Joseph K. W. Fung 12.3

Prof. Joseph K. W. Fung 12.4

 Su = Su = $22, u = 1.1 (up-jump factor)

 Sd = Sd = $18, d = 0.9 (down-jump factor)

Prof. Joseph K. W. Fung 12.5

A 3-month call option on the stock has a strike price of $21

Prof. Joseph K. W. Fung 12.6

 Portfolio is “risk-free” when $22– $1 = $18 or  = 0.25

Prof. Joseph K. W. Fung 12.7

Short 0.25 Shares of Stock +$5 – $5.5 – $4.5

A derivative ( f ) lasts for time T and is

Prof. Joseph K. W. Fung 12.13

 The portfolio is risk-free when Su– ƒu = Sd– ƒd or

 Since the portfolio is risk-free, value of the portfolio today is

 The value of the portfolio today is S– f = (Su –

 Hence, ƒ = S– (Su– ƒu )e–rT

Prof. Joseph K. W. Fung 12.15

Prof. Joseph K. W. Fung 12.16

Prof. Joseph K. W. Fung 12.17

 RHS: 1st term is a stock position and 2nd term

Prof. Joseph K. W. Fung 12.18

ƒ = Ep (f) e–rT = [p ƒu + (1 – p )ƒd ]e–rT

Prof. Joseph K. W. Fung 12.20

Investors are risk neutral if

Hence p & 1 – p are called risk-neutral probabilities

The value of the option is

= e–0.120.25 (0.65231 + 0.34770) = $0.633

Note that the pricing formula fails to show how to trade

Prof. Joseph K. W. Fung 12.25

Δ = 0.25  (long 0.25 shares per 1 short call)

 If stock price goes up, value of portfolio = $22(0.25) – $1 = $4.5

 If stock price drops, value of portfolio = $18(0.25) = $4.5

 Present value of portfolio = $4.5e-0.12/4 = $4.367

Prof. Joseph K. W. Fung 12.26

Δ = (fu – fd) / (Su – Sd)

Δ: ratio of change in option (any other derivative)

Delta shows how the value of a derivative is affected

S = $20 Sud = $19.8

 Each time step is 3 months; total length 6 m.

 Value at node B = e–0.120.25(0.65233.2 + 0.34770) =

Prof. Joseph K. W. Fung 12.29

fu = [ pfuu + (1 – p)fud ] e-rT

fd = [ pfud + (1 – p)fdd ] e-rT

f = [ pfu + (1 – p)fd ] e-rT

f = e-r2T [ p2fuu + 2p(1 – p)fud + (1 – p)2fdd ]

Prof. Joseph K. W. Fung 12.31

j!n  j ! is the (binomial) probability that the

value of the call is equal to

periods or (n moves) when the option expires

u a d na S  X and the call finishes in-the-money,

Prof. Joseph K. W. Fung 12.33

a is the smallest non-negative integer greater than

Prof. Joseph K. W. Fung 12.34

Prof. Joseph K. W. Fung 12.35

The first bracketed term can be simplified as

Prof. Joseph K. W. Fung 12.36

Prof. Joseph K. W. Fung 12.37

a; n; p  is the probability of exercise

If S is large relative to X, a is small and the probability of