Class 8 Dynamic Replication

Model Description Single-Period tree Multi-period tree Calibration Extensions Trinomial Trees
Dynamic Replication
Mario Fuenzalida Carlos Lafuente Marcelo Ortiz
Universidad Adolfo Ibanez
April 25, 2011
Fuenzalida, Lafuente, Ortiz
Dynamic Replication
Contents
1
Model Description
Dynamic Replication
Contents
1
Model Description Risky and risk-free asset
Dynamic Replication
Contents
1
Model Description Risky and risk-free asset Relation between u, d, and r .
Dynamic Replication
Contents
1 2
Model Description Risky and risk-free asset Relation between u, d, and r . Single-Period tree
Dynamic Replication
Contents
1 2
Model Description Risky and risk-free asset Relation between u, d, and r . Single-Period tree General Derivative Pay-o
Dynamic Replication
Contents
1 2
Model Description Risky and risk-free asset Relation between u, d, and r . Single-Period tree General Derivative Pay-o Pay-o replication Condition
Dynamic Replication
Contents
1 2
Model Description Risky and risk-free asset Relation between u, d, and r . Single-Period tree General Derivative Pay-o Pay-o replication Condition Replication portfolio Composition
Dynamic Replication
Contents
1 2
Risk-Neutral probabilities
Dynamic Replication
Contents
1 2
Risk-Neutral probabilities Delta
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities Calibration
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities Calibration Risk-Neutral stock dynamics
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities Calibration Risk-Neutral stock dynamics Denition of Volatility
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities Calibration Risk-Neutral stock dynamics Denition of Volatility Moment matching of normal instantaneous returns
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities Calibration Risk-Neutral stock dynamics Denition of Volatility Moment matching of normal instantaneous returns Expression of u and d
Dynamic Replication
Contents
1 2
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities Calibration Risk-Neutral stock dynamics Denition of Volatility Moment matching of normal instantaneous returns Expression of u and d Extensions
Fuenzalida, Lafuente, Ortiz Dynamic Replication
Contents
1 2
5 6
Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities Calibration Risk-Neutral stock dynamics Denition of Volatility Moment matching of normal instantaneous returns Expression of u and d Extensions Trinomial Trees
Risky and risk-free asset Relation between u, d, and r .
Model Description
Suppose you have a risk-free asset, like a MMA (money market account) in wich you make a deposit of M USD. The rate oered by the bank at moment t0 is rt 0.
Suppose now, that you have a risky asset, like a stock , that at t0 have a price of S0 The stock price can either move up from S0 to u S0 , where u > 1, or down from S0 to d S0 , where d < 1, and u = 1/d.
Dynamic Replication
Risky and risk-free asset Relation between u, d, and r .
Obviously, (1 + r ) must be lower than u, because in other way invest in risky asset (like stock) will be less protable than invest in risk free asset, even in the best possible return that risky asset can generate. So: (1 + r ) < u
Dynamic Replication
General Derivative Pay-o Pay-o replication Condition Replication portfolio Composition Arrow-Debreu Securities
Dynamic Replication
General derivative Pay-o Option European Vanilla Call
Dynamic Replication
Dynamic Replication
General derivative Pay-o Option European Vanilla Call No arbitrage
Dynamic Replication
Dynamic Replication
General derivative Pay-o Option European Vanilla Call No arbitrage Option can only be exercised at the end of its life, at its maturity.
Dynamic Replication
Dynamic Replication
General derivative Pay-o Option European Vanilla Call No arbitrage Option can only be exercised at the end of its life, at its maturity. At t1 the pay-o of the option is: Max (St1 -K,0)
Dynamic Replication
Dynamic Replication
General derivative Pay-o Option European Vanilla Call No arbitrage Option can only be exercised at the end of its life, at its maturity. At t1 the pay-o of the option is: Max (St1 -K,0) We need calculate option value at t0 : f0 .
Dynamic Replication
Dynamic Replication
General derivative Pay-o Option European Vanilla Call No arbitrage Option can only be exercised at the end of its life, at its maturity. At t1 the pay-o of the option is: Max (St1 -K,0) We need calculate option value at t0 : f0 . How can we nd f0 ? Dynamic Replication
Dynamic Replication
Dynamic Replication
General Derivative Pay-o Replicate the option.
Dynamic Replication
Dynamic Replication
General Derivative Pay-o Replicate the option. Bonds bought for the price 1 and stocks.
Dynamic Replication
Dynamic Replication
General Derivative Pay-o Replicate the option. Bonds bought for the price 1 and stocks. At t0 , F0 = + S0
Dynamic Replication
Dynamic Replication
Pay-o replication Condition At t1 there are two possibilities: Fu = (1 + r ) + S0 u or Fd = (1 + r ) + S0 d
Dynamic Replication
Dynamic Replication
Pay-o replication Condition At t1 there are two possibilities: Fu = (1 + r ) + S0 u or Fd = (1 + r ) + S0 d Solve the system of equations... F u Fd S0 (u d) uFd Fu d = R(u d) =
Dynamic Replication
Dynamic Replication
Replication portfolio Composition We will illustrate the concepts with a European vanilla call option with strike K = 100 on a non-dividend paying stock with initial value S0 = 100, with a risk-free interest rate r = 0, and model parameters u = 1,2 and d = 0,8.
Figure: Single-PeriodReplication tree Dynamic
Dynamic Replication
Replication portfolio Composition Replacing... F u Fd S0 (u d) uFd Fu d = R(u d) =
Dynamic Replication
Dynamic Replication
Replication portfolio Composition Replacing... F u Fd S0 (u d) uFd Fu d = R(u d) = = 20 0 = 100(1, 2 0, 8)
1 2
Dynamic Replication
Dynamic Replication
Replication portfolio Composition Replacing... F u Fd S0 (u d) uFd Fu d = R(u d) = 20 0 =1 2 100(1, 2 0, 8) 1, 2 0 20 0, 8 = = 40 1(1, 2 0, 8) =
Dynamic Replication
Dynamic Replication
Replication portfolio Composition Replacing... F u Fd S0 (u d) uFd Fu d = R(u d) = 20 0 =1 2 100(1, 2 0, 8) 1, 2 0 20 0, 8 = = 40 1(1, 2 0, 8) Thus, we need to borrow 40 dollars and then buy 0.5 units of stock at t0 . The cost of this will be the current value of the option: = F0 =
1 2
100 40 = 10
Dynamic Replication
Replication Portfolio Condition The variable p is the probability of a up movement, then, 1-p, is the probability of a down movement, and the expected stock price at time T, E(ST ), is given by: E(ST ) = pS0 u + (1-p)S0 d E(ST ) = pS0 (u-d) + S0 d
Dynamic Replication
Dynamic Replication
Replication Portfolio Condition The variable p is the probability of a up movement, then, 1-p, is the probability of a down movement, and the expected stock price at time T, E(ST ), is given by: E(ST ) = pS0 u + (1-p)S0 d E(ST ) = pS0 (u-d) + S0 d But the stock price grows on average at the risk-free rate E(ST ) = S0 erT
Dynamic Replication
Dynamic Replication
Replication Portfolio Condition We now return to the example, and illustrate that risk-neutral valuation. The stock price is currently $100 and will move either up to $120 or down to $80. The option considered is a European call option with a strike price of $100.
Dynamic Replication
Dynamic Replication
Replication Portfolio Condition We now return to the example, and illustrate that risk-neutral valuation. The stock price is currently $100 and will move either up to $120 or down to $80. The option considered is a European call option with a strike price of $100. We dene p as the probability of an upward movement in the stock price in risk-neutral world. We can calculate p:
Dynamic Replication
Dynamic Replication
Replication Portfolio Condition We now return to the example, and illustrate that risk-neutral valuation. The stock price is currently $100 and will move either up to $120 or down to $80. The option considered is a European call option with a strike price of $100. We dene p as the probability of an upward movement in the stock price in risk-neutral world. We can calculate p: 100 = 100p - 80(1-p) p = 0,5
Dynamic Replication
Dynamic Replication
Replication portfolio Condition Delta is an important parameter in the pricing and hedging of options.
Dynamic Replication
Dynamic Replication
Replication portfolio Condition Delta is an important parameter in the pricing and hedging of options. The delta of a stock option is the ratio of the change in the price of the stock option to the change in the price of the underlying stock.
Dynamic Replication
Dynamic Replication
Replication portfolio Condition Delta is an important parameter in the pricing and hedging of options. The delta of a stock option is the ratio of the change in the price of the stock option to the change in the price of the underlying stock. It is the number of units of the stock we should hold for each option shorted in order to create a riskless hedge.
Dynamic Replication
Dynamic Replication
Replication portfolio Condition Delta is an important parameter in the pricing and hedging of options. The delta of a stock option is the ratio of the change in the price of the stock option to the change in the price of the underlying stock. It is the number of units of the stock we should hold for each option shorted in order to create a riskless hedge. In the example:
200 12080
1 2
Dynamic Replication
Dynamic Replication
Replication portfolio Condition Delta is an important parameter in the pricing and hedging of options. The delta of a stock option is the ratio of the change in the price of the stock option to the change in the price of the underlying stock. It is the number of units of the stock we should hold for each option shorted in order to create a riskless hedge. In the example:
200 12080
1 2
This is because when the stock price changes from $80 to $120, the option price changes from $0 to $20.
Dynamic Replication
Dynamic Replication
Arrow-Debreu Securities Can be understood as Arrow-Debreu assets how insurance contracted for a certain amount of money if a certain state of nature occurs.
1 1 + 2 = 1+r = b uS0 1 + dS0 2 = S0 F0 = Fu 1 + Fd 2
Dynamic Replication
Dynamic Replication
1 1 + 2 = 1+r = b uS0 1 + dS0 2 = S0 F0 = Fu 1 + Fd 2
where r is the return on risk-free asset and b is the present value of a bond that pays $1 in one year.
Dynamic Replication
Dynamic Replication
1 1 + 2 = 1+r = b uS0 1 + dS0 2 = S0 F0 = Fu 1 + Fd 2
where r is the return on risk-free asset and b is the present value of a bond that pays $1 in one year. 1 price of the Arrow-Debreu asset 1.
Dynamic Replication
Dynamic Replication
1 1 + 2 = 1+r = b uS0 1 + dS0 2 = S0 F0 = Fu 1 + Fd 2
where r is the return on risk-free asset and b is the present value of a bond that pays $1 in one year. 1 price of the Arrow-Debreu asset 1. 2 price of the Arrow-Debreu asset 2.
Dynamic Replication
Dynamic Replication
Arrow-Debreu Securities
Dynamic Replication
Dynamic Replication
Arrow-Debreu Securities Example: S0 =100, uS0 =120, dS0 =80,
1 1+r =1,
Fu =20, Fd =0
Dynamic Replication
Dynamic Replication
Arrow-Debreu Securities Example: S0 =100, uS0 =120, dS0 =80, 1 = ?, 2 = ?, F0 = ?
1 1+r =1,
Fu =20, Fd =0
Dynamic Replication
Dynamic Replication
Arrow-Debreu Securities Example: S0 =100, uS0 =120, dS0 =80, 1 = ?, 2 = ?, F0 = ? Replacing... 1201 + 802 = 100 1 + 2 = 1
1 1+r =1,
Fu =20, Fd =0
Dynamic Replication
Dynamic Replication
Arrow-Debreu Securities Example: S0 =100, uS0 =120, dS0 =80, 1 = ?, 2 = ?, F0 = ? Replacing... 1201 + 802 = 100 1 + 2 = 1 Resolving: 1 = 0,5, 2 = 0,5
1 1+r =1,
Fu =20, Fd =0
Dynamic Replication
Dynamic Replication
Arrow-Debreu Securities Example: S0 =100, uS0 =120, dS0 =80, 1 = ?, 2 = ?, F0 = ? Replacing... 1201 + 802 = 100 1 + 2 = 1 Resolving: 1 = 0,5, 2 = 0,5 thus, the price of option is: 20*0,5 + 0*0,5 = 10
1 1+r =1,
Fu =20, Fd =0
Dynamic Replication
Dynamic Replication
Arrow-Debreu Securities Example: S0 =100, uS0 =120, dS0 =80, 1 = ?, 2 = ?, F0 = ? Replacing... 1201 + 802 = 100 1 + 2 = 1 Resolving: 1 = 0,5, 2 = 0,5 thus, the price of option is: 20*0,5 + 0*0,5 = 10 the same previous result.
1 1+r =1,
Fu =20, Fd =0
Convexity: Impossibility of static replication Backward replication Prices of the Arrow-Debreu securities Risk-neutral probabilities
Replicating portfolios may need constant adjustment to maintain their equivalence with the targeted instrument: 1) if markets in the component instruments do not exist 2) the instruments themselves may exist, but they may not be liquid. 3) the asset to be replicated can be highly nonlinear
Principles of dynamic replication: The strategy will combine imperfect instruments that are correlated with each other to get a synthetic at time t0 . If these price of that instruments (random variables) were correlated in a certain fashion, these correlations can be used against each other to eliminate the need for cash injections or withdrawals. The cost of forming the portfolio at t0 would then equal the arbitrage-free value of the original asset.
Dynamic Replication
Obviously, if the instruments are time-independent, their correlations are 0, so, they cant eliminate the need of cash injections or withdrawals. If we want to replicate the payo of a call at time t < T , we must look that their value is nonlinear.
Obviously, is impossible replicate their value using time-independent instrument.

Is a procedure for working from the end of a tree to its beginning, in order to value the weight of the replications instrument. Example: We want replicate a Bond with maturity T2 (the value of the bond at T2 is $100). Suppose the market practitioner has only two liquid markets: 1)The rst is the market for one-period lending/borrowing, denoted by the symbol Bt . The Bt is the time t value of $1 invested at time t0 . 2) The second liquid market is for a default-free pure discount bond whose time-t price is denoted by B(t, T3 ). The bond pays $100 at time T3 and sells for the price B(t, T3 ) at time t. The idea is to use the information given in trees of the lending-borrowing lend and the bond, to form a portfolio with (time-varying) weights t and bond t for Bt and B(t, T3 ).
Dynamic Replication
The portfolio should mimic the value of the medium-term bond B(t, T2 ) at all nodes at t = 0, 1, 2. The rst condition on this portfolio is that, at T2 , its value must equal $100.
lend bond This means that the t and t will satisfy: lend bond t Bt + t B(t, T3 ) = B(t, T2 )
, for all t. lend bond Like we want that adjustments of the weights t and t , observed along the tree paths t = 0, 1, 2 will not lead to any cash injections or withdrawals, so:
Dynamic Replication
lend up bond lend up bond t Bt+1 + t B(t + 1, T3 )up = t+1 Bt+1 + t+1 B(t + 1, T3 )up lend down bond lend down bond t Bt+1 + t B(t + 1, T3 )down = t+1 Bt+1 + t+1 B(t + 1, T3 )down
The left-hand side is the value of a portfolio chosen at time t, and valued at a new up or down state at time t + 1. The right-hand side represents the cost of a new portfolio chosen at time t + 1, either in the up or down state. Putting these two together, the equations imply that, regardless of which state occurs, the previously chosen portfolio generates just enough cash to put together a new replicating portfolio
Dynamic Replication
Imagine that are just two possible states tomorrow: Up (u) and Down (d). Denote the random variable which represents the state as S; denote tomorrows random variable as S1 . Thus, S1 can take two values: S1 = u and S1 = d. Suppose that: 1)Exist a security that pays o $1 if tomorrows state is u and $0 if state is d. The price of this security is pu . 2)Exist other security that pays o $1 if tomorrows state is d and nothing if the state is u. The price of this security is pd . The prices pd and pu are the state prices. Theprobabilitiesof S1 = u and S1 = d aect to state prices. The more likely a move to d is, the higher the price pd gets, since pd insures the agent against the occurrence of state d. The seller of this insurance would demand a higher premium.
Again, suppose that are just two possible states tomorrow: Up (u) and Down (d). Dene a Price Vector, like a vector composed with the currents prices of 3 instruments: risk-free deposit of B USD, an asset valued on P USD, and a their Call Option of c USD. Dene now a Payo Matrix , wich include the two possible payo, depending of the states, for each instruments. In a free arbitrage market, there are two positive constraints ,u ,d allowing the next equation: B B(1 + r ) B(1 + r ) P = Pu Pd u (1) d c cu cd
Dynamic Replication
Naturally, these constraints are the Arrow-Debreu securities. Dividing the rst equation by B, we obtain: 1 = (1 + r )u + (1 + r )d We can re-write the last equation using the next substitution:
u = (1 + r )u d = (1 + r )d 1 = u + d
We can conclude that risk-neutral probabilities are nothing but the Arrow-Debreu prices in disguise. Increasing the number of steps N (each step with length dT = T /N), asymptotically, the distribution becomes normal.
Risk-Neutral stock dynamics Denition of Volatility Moment matching of normal instantaneous returns Expression of u and d
Dynamic Replication
Risk-Neutral stock dynamics The model of stock price behavior used by Black, Scholes and Merton assumes that percentage changes in the stock price in a short period of time are normally distributed, where:
Dynamic Replication
Dynamic Replication
Risk-Neutral stock dynamics The model of stock price behavior used by Black, Scholes and Merton assumes that percentage changes in the stock price in a short period of time are normally distributed, where: = Expected return on stock per year.
Dynamic Replication
Dynamic Replication
Risk-Neutral stock dynamics The model of stock price behavior used by Black, Scholes and Merton assumes that percentage changes in the stock price in a short period of time are normally distributed, where: = Expected return on stock per year. = Volatility of the stock price per year.
Dynamic Replication
Dynamic Replication
Risk-Neutral stock dynamics The model of stock price behavior used by Black, Scholes and Merton assumes that percentage changes in the stock price in a short period of time are normally distributed, where: = Expected return on stock per year. = Volatility of the stock price per year. Standard deviation of this percentage change is t so that: S S ( t, t )
Dynamic Replication
Dynamic Replication
Risk-Neutral stock dynamics The model of stock price behavior used by Black, Scholes and Merton assumes that percentage changes in the stock price in a short period of time are normally distributed, where: = Expected return on stock per year. = Volatility of the stock price per year. Standard deviation of this percentage change is t so that: S S ( t, t ) Where S is the change in the stock price S in time t, and (m, s) denotes a normal distribution with mean m and standard deviation s.
Dynamic Replication
Dynamic Replication
Risk-Neutral stock dynamics From Itos lemma dG = ( G S + S
G t
1 2G 2 2 S )dt + 2 S 2
G S Sdz
Dynamic Replication
Dynamic Replication
Risk-Neutral stock dynamics From Itos lemma 1 2G 2 2 S )dt + 2 S 2 Where G = ln(S), and constant dG = ( G S + S
G t
G S Sdz
Dynamic Replication
Dynamic Replication
G t
G S Sdz
This means that: ln(ST ) - ln(S0 ) [(

2 2 ),
T]
Dynamic Replication
Dynamic Replication
G t
G S Sdz
This means that: ln(ST ) - ln(S0 ) [( From this, it follows that:

2 2 ),
T]
2 ln( ST ) [( ), T ] S0 2 and 2 ln(ST ) [ln(S0 ) + ( ), T ] 2
Dynamic Replication
Dynamic Replication
Denition of Volatility The volatility of a stock price can be dened as the standard deviation of the return provided by the stock in 1 year when the return is expressed using continuous compounding.
Dynamic Replication
Dynamic Replication
Denition of Volatility The volatility of a stock price can be dened as the standard deviation of the return provided by the stock in 1 year when the return is expressed using continuous compounding. The volatility of a stock is a measure of our uncertainly about the returns provided by the stock. Stocks typically have a volatility between 15% and 60%.
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns In practice, when constructing a binomial tree to represent the movements in a stock price, we choose the parameters u and d to match the volatility of the stock price.
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns In practice, when constructing a binomial tree to represent the movements in a stock price, we choose the parameters u and d to match the volatility of the stock price. To see how this is done, we suppose that the expected return on a stock is u and its volatility is .
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns The probability of an up movement is assumed to be p.
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns The probability of an up movement is assumed to be p. The expected stock price at the end of rst time step is S0 e ut . On the tree the expected stock price at this time is: pS0 u + (1 - p)S0 d
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns The probability of an up movement is assumed to be p. The expected stock price at the end of rst time step is S0 e ut . On the tree the expected stock price at this time is: pS0 u + (1 - p)S0 d In order to match the expected return on the stock with the trees parameters, we must therefore have: pS0 u + (1 - p)S0 d = S0 e ut
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns The volatility of a stock price is dened so that t, that is the standard deviation of the return on the stock price in a short period of time of lenght t.
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns The volatility of a stock price is dened so that t, that is the standard deviation of the return on the stock price in a short period of time of lenght t. Equivalently, the variance of the return is 2 t.
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns The volatility of a stock price is dened so that t, that is the standard deviation of the return on the stock price in a short period of time of lenght t. Equivalently, the variance of the return is 2 t. Variance of X is: E(X 2 ) - [E (X )]2
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns The volatility of a stock price is dened so that t, that is the standard deviation of the return on the stock price in a short period of time of lenght t. Equivalently, the variance of the return is 2 t. Variance of X is: E(X 2 ) - [E (X )]2 Then: pu 2 + (1-p)d 2 - [pu + (1 p)d]2
Dynamic Replication
Dynamic Replication
Moment matching of normal instantaneous returns The volatility of a stock price is dened so that t, that is the standard deviation of the return on the stock price in a short period of time of lenght t. Equivalently, the variance of the return is 2 t. Variance of X is: E(X 2 ) - [E (X )]2 Then: pu 2 + (1-p)d 2 - [pu + (1 p)d]2 In order to match the stock price volatility with the trees parameters, we must therefore have: pu 2 + (1-p)d 2 - [pu + (1 p)d]2 = 2 t
Dynamic Replication
Dynamic Replication
Expression of u and d Substituting: e ut (u+d) - ud - e 2ut = 2 t
Dynamic Replication
Dynamic Replication
Expression of u and d Substituting: e ut (u+d) - ud - e 2ut = 2 t Using the series expansion: e x = 1 + x +

t u = e t d=e x2 2 x3 3
+ ...
Dynamic Replication
Dynamic Replication
Expression of u and d Substituting: e ut (u+d) - ud - e 2ut = 2 t Using the series expansion: e x = 1 + x +

t u = e t d=e x2 2 x3 3
+ ...
These are the evalues of u and d proposed by Cox, Ross and Rubinstein (1979) for matching u and d.
Dynamic Replication
Dynamic Replication
Early exercise:
Dynamic Replication
Dynamic Replication
Early exercise: An early exercise of an American option will not realized at any moment of the life of the contract if: K > S(t), but if S(t) > K may not be exercised if the investor thinks that the price of the stock will be higher.
Dynamic Replication
Dynamic Replication
Early exercise: An early exercise of an American option will not realized at any moment of the life of the contract if: K > S(t), but if S(t) > K may not be exercised if the investor thinks that the price of the stock will be higher. Pricing American options is more complex than the pricing of European options since it is required an optimization, related to the optimal time for the investor for exercising.
Dynamic Replication
Dynamic Replication
Early exercise: An early exercise of an American option will not realized at any moment of the life of the contract if: K > S(t), but if S(t) > K may not be exercised if the investor thinks that the price of the stock will be higher. Pricing American options is more complex than the pricing of European options since it is required an optimization, related to the optimal time for the investor for exercising. EEP is the early exercise premium which expresses the value of the American option as the corresponding European option value plus the gain from early exercise which is the present value of the dividend benets in the exercise region net of the interest losses on the payments incurred upon exercise.
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock It is never optimal to exercise an American call option over a NDPS if the investor plans to keep the stock: T = 1 month; SK =40;S(t)=50
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock It is never optimal to exercise an American call option over a NDPS if the investor plans to keep the stock: T = 1 month; SK =40;S(t)=50 If the investor wants to hold the stock after the contract he will pay USD 40 one month later, and invest that value for the entire month making interest for that period, the stock pays no dividends.
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock It is never optimal to exercise an American call option over a NDPS if the investor plans to keep the stock: T = 1 month; SK =40;S(t)=50 If the investor wants to hold the stock after the contract he will pay USD 40 one month later, and invest that value for the entire month making interest for that period, the stock pays no dividends. Also there is some probability that the stock price in one more month may be lower than USD 40.
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock It is never optimal to exercise an American call option over a NDPS if the investor plans to keep the stock: T = 1 month; SK =40;S(t)=50 If the investor wants to hold the stock after the contract he will pay USD 40 one month later, and invest that value for the entire month making interest for that period, the stock pays no dividends. Also there is some probability that the stock price in one more month may be lower than USD 40. If he thinks the stock is overvalued is better to sell the option or short the stock and keep the option, so the price obtained will be grater than the intrinsic value of S(t) K = 10
Dynamic Replication
Dynamic Replication
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock The value of and option must satises that: c >= S0 Ke rt
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock The value of and option must satises that: c >= S0 Ke rt Because in an American call option the investor has all the period to decide when to exercise the contract, its value is grater than of an equivalence European call option: C >= c;C >= S0 Ke rt
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock The value of and option must satises that: c >= S0 Ke rt Because in an American call option the investor has all the period to decide when to exercise the contract, its value is grater than of an equivalence European call option: C >= c;C >= S0 Ke rt Since r > 0, then: C > S0 K , and if it were optimal to exercise before the contract gets the maturity it is required that: C = S0 K .
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a non-dividend-paying stock The value of and option must satises that: c >= S0 Ke rt Because in an American call option the investor has all the period to decide when to exercise the contract, its value is grater than of an equivalence European call option: C >= c;C >= S0 Ke rt Since r > 0, then: C > S0 K , and if it were optimal to exercise before the contract gets the maturity it is required that: C = S0 K . The value of a call is always above its intrinsic value, max(S0 K , 0), and while r, T and the volatility increases, the call price also increase.
Dynamic Replication
Dynamic Replication
Dynamic Replication
Dynamic Replication
Holding the option until maturity gives a protection to the investor in case the stock prices gets lower than the strike price (insurance).
Dynamic Replication
Holding the option until maturity gives a protection to the investor in case the stock prices gets lower than the strike price (insurance). Time value of money, is better to pay latter than sooner.
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock It is optimal to exercise an American call option over a DPS only at a time immediately before the stock goes ex-dividend.
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock It is optimal to exercise an American call option over a DPS only at a time immediately before the stock goes ex-dividend. Assume that remains n ex-dividend dates at time t1 , t2 , tn and their present values are D1 , D2 , Dn . The early exercise will be considered for the tn , the investor will receive S(tn ) K if he decided to exercise the option.
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock It is optimal to exercise an American call option over a DPS only at a time immediately before the stock goes ex-dividend. Assume that remains n ex-dividend dates at time t1 , t2 , tn and their present values are D1 , D2 , Dn . The early exercise will be considered for the tn , the investor will receive S(tn ) K if he decided to exercise the option. The lower bound for an American call on a DPS is: C >= S0 D Ke rT
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock It is optimal to exercise an American call option over a DPS only at a time immediately before the stock goes ex-dividend. Assume that remains n ex-dividend dates at time t1 , t2 , tn and their present values are D1 , D2 , Dn . The early exercise will be considered for the tn , the investor will receive S(tn ) K if he decided to exercise the option. The lower bound for an American call on a DPS is: C >= S0 D Ke rT For the specic case shown: C >= S(tn ) Dn Ke r (T tn ) .
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock It is optimal to exercise an American call option over a DPS only at a time immediately before the stock goes ex-dividend. Assume that remains n ex-dividend dates at time t1 , t2 , tn and their present values are D1 , D2 , Dn . The early exercise will be considered for the tn , the investor will receive S(tn ) K if he decided to exercise the option. The lower bound for an American call on a DPS is: C >= S0 D Ke rT For the specic case shown: C >= S(tn ) Dn Ke r (T tn ) . Then: S(tn ) Dn Ke r (T tn ) > S(tn ) K , making Dn <= K [1 e r (T tn ) ].
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock It is optimal to exercise an American call option over a DPS only at a time immediately before the stock goes ex-dividend. Assume that remains n ex-dividend dates at time t1 , t2 , tn and their present values are D1 , D2 , Dn . The early exercise will be considered for the tn , the investor will receive S(tn ) K if he decided to exercise the option. The lower bound for an American call on a DPS is: C >= S0 D Ke rT For the specic case shown: C >= S(tn ) Dn Ke r (T tn ) . Then: S(tn ) Dn Ke r (T tn ) > S(tn ) K , making Dn <= K [1 e r (T tn ) ]. In that case is not optimal to early exercise the option at time tn .
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock On the opposite way if: Dn >= K [1 e r (T tn ) ].
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock On the opposite way if: Dn >= K [1 e r (T tn ) ]. It is always optimal to exercise the option if the stock value at time n is suciently high. The inequality will be satised if n is close to the expiry date and the dividend is large.
Dynamic Replication
Dynamic Replication
Early exercise: : Calls on a dividend-paying stock On the opposite way if: Dn >= K [1 e r (T tn ) ]. It is always optimal to exercise the option if the stock value at time n is suciently high. The inequality will be satised if n is close to the expiry date and the dividend is large. The relation between the dividend and the price of the stock is called dividend yield.
Dynamic Replication
Dynamic Replication
Dividend Yield
Dynamic Replication
Dynamic Replication
Dividend Yield The relation between the dividend and the price of the stock is called dividend yield. This ratioshows how much a company pays out in dividends each year relative to its stock price. Isa way to measure how much cash ow is getting for each dollar invested in an equity position.
Dynamic Replication
Dynamic Replication
Dividend Yield The relation between the dividend and the price of the stock is called dividend yield. This ratioshows how much a company pays out in dividends each year relative to its stock price. Isa way to measure how much cash ow is getting for each dollar invested in an equity position. Calculation is as follows:
Annualdividendsperstock Stockprice
Dynamic Replication
Dynamic Replication
Discrete dividends
Dynamic Replication
Dynamic Replication
Discrete dividends In the Black-Scholes model, any dividends on stocks are paid continuously, but in reality dividends are always paid discretely.
Dynamic Replication
Dynamic Replication
Discrete dividends In the Black-Scholes model, any dividends on stocks are paid continuously, but in reality dividends are always paid discretely. It is not entirely clear how such discrete dividends are to be handled; simple perturbations of the Black-Scholes model often fall into contradictions.
Dynamic Replication
Dynamic Replication
Discrete dividends In the Black-Scholes model, any dividends on stocks are paid continuously, but in reality dividends are always paid discretely. It is not entirely clear how such discrete dividends are to be handled; simple perturbations of the Black-Scholes model often fall into contradictions. The market convention is to recognize the stock price as the net present value of all future dividends, and to model the (discrete) dividend process directly.
Dynamic Replication
Dynamic Replication
Discrete dividends In the Black-Scholes model, any dividends on stocks are paid continuously, but in reality dividends are always paid discretely. It is not entirely clear how such discrete dividends are to be handled; simple perturbations of the Black-Scholes model often fall into contradictions. The market convention is to recognize the stock price as the net present value of all future dividends, and to model the (discrete) dividend process directly. Arbitrage pricing theory (APT) states the price of a stock as the present value of the future dividend payments.
Dynamic Replication
Dynamic Replication
Discrete dividends In the Black-Scholes model, any dividends on stocks are paid continuously, but in reality dividends are always paid discretely. It is not entirely clear how such discrete dividends are to be handled; simple perturbations of the Black-Scholes model often fall into contradictions. The market convention is to recognize the stock price as the net present value of all future dividends, and to model the (discrete) dividend process directly. Arbitrage pricing theory (APT) states the price of a stock as the present value of the future dividend payments. The problem is to write down the joint distribution of the two random variables, dividends and stocks price, because sometimes leads to an inconsistency.
Dynamic Replication
Discrete dividends
Dynamic Replication
Dynamic Replication
Discrete dividends Modeling the stock price with a discrete set of dividends payments at dates t1 < t2 , is should be useful to modeling the dividend payments, is then a consequence of the model assumed for the dividends the stock price process.
Dynamic Replication
Dynamic Replication
Replacing the stock with a future
Dynamic Replication
Dynamic Replication
Replacing the stock with a future One way to do this is using a Total Return Swap, which is an agreement to exchange the total return of an asset for LIBOR plus a spread. The total returns involves all the cash ows related to the asset, for the case of a stock will pays dividends and capital gains, the receiver pays interest at LIBOR plus a spread, both based on a principal amount.
Dynamic Replication
Dynamic Replication
Replacing the stock with a future One way to do this is using a Total Return Swap, which is an agreement to exchange the total return of an asset for LIBOR plus a spread. The total returns involves all the cash ows related to the asset, for the case of a stock will pays dividends and capital gains, the receiver pays interest at LIBOR plus a spread, both based on a principal amount. The LIBOR is set for next period at the last dividend payment date.
Dynamic Replication
Dynamic Replication
Replacing the stock with a future One way to do this is using a Total Return Swap, which is an agreement to exchange the total return of an asset for LIBOR plus a spread. The total returns involves all the cash ows related to the asset, for the case of a stock will pays dividends and capital gains, the receiver pays interest at LIBOR plus a spread, both based on a principal amount. The LIBOR is set for next period at the last dividend payment date. When the contract gets it maturity, there is a payment for compensating the change in the stock value.
Dynamic Replication
Dynamic Replication
Dynamic Replication
Dynamic Replication
Other way is to use a Dividend Swap, in which investor can buy or sell the dividends paid by a stock or an index. They can be used for hedging or managing cash ows from stocks portfolios.
Dynamic Replication
Dynamic Replication
Other way is to use a Dividend Swap, in which investor can buy or sell the dividends paid by a stock or an index. They can be used for hedging or managing cash ows from stocks portfolios. Dividend Swap, compared to the Total Return Swap, exchange xed by oating amounts, were the oating is related to the dividends made.
Dynamic Replication
Other way is to use a Dividend Swap, in which investor can buy or sell the dividends paid by a stock or an index. They can be used for hedging or managing cash ows from stocks portfolios. Dividend Swap, compared to the Total Return Swap, exchange xed by oating amounts, were the oating is related to the dividends made. Their payouts are determined by the dividends reach by the stock, Fuenzalida, Lafuente, Ortiz Dynamic Replication
Dynamic Replication
Construction - General aspects
Dynamic Replication
Dynamic Replication
Construction - General aspects Improves upon the binomial model by allowing a stock price to move up, down or stay the same with certain probabilities, as shown in the diagram below:
Dynamic Replication
Dynamic Replication
Construction - General aspects Improves upon the binomial model by allowing a stock price to move up, down or stay the same with certain probabilities, as shown in the diagram below:
They are considered an eective method of numerical calculation of option prices within B-S model.
Dynamic Replication
Construction - Model denition
Dynamic Replication
Dynamic Replication
Construction - Model denition The price of a stock in the next period under the trinomial tree could have the following values:
Dynamic Replication
Dynamic Replication
Construction - Model denition The price of a stock in the next period under the trinomial tree could have the following values: S(t + t) = S(t)u with probability pu
Dynamic Replication
Dynamic Replication
Construction - Model denition The price of a stock in the next period under the trinomial tree could have the following values: S(t + t) = S(t)u with probability pu S(t + t) = S(t) with probability 1 pu pd
Dynamic Replication
Dynamic Replication
Construction - Model denition The price of a stock in the next period under the trinomial tree could have the following values: S(t + t) = S(t)u with probability pu S(t + t) = S(t) with probability 1 pu pd S(t + t) = S(t)d with probability pd
Dynamic Replication
Dynamic Replication
Dynamic Replication
Dynamic Replication
Construction - Model denition Matching the rst two moments of the models distribution to the no arbitrage condition: E [S(ti+1 |S(ti ] = e r t S(ti ) VAR[S(ti+1 |S(ti ] = t S(ti )2 2 + o(t) Assuming constant volatility and that the stock price follows a geometric Brownian motion. The rst condition states an equilibrium assumption where the average return of the stock should be equal to the risk free return. Also it is required that the upward jump most be identical to the downward jump.
Dynamic Replication
Dynamic Replication
Dynamic Replication
Dynamic Replication
Construction - Model denition Having the values of the jump sizes u and d, and the transition probabilities pu and pd , it is possible to reach the value of the stock for any sequence of price movements. Other thing to specify is the number of jumps Nu , Nd and Nm , after that it is possible to reach the stock price at node j for time i as: Si,j = u Nu d Nd S(t0 ), where Nu + Nd + Nm = n
Dynamic Replication
Dynamic Replication
Construction - Setting the parameters
Dynamic Replication
Dynamic Replication
Construction - Setting the parameters The sizes of the jumps are: u = e
2t
, d = e
2t
The transition probabilities are given by: pu = e (r t)/2 e (t)/2 e (t)/2 e (t)/2 e (t)/2 e (r t)/2 e (t)/2 e (t)/2 pm = 1 pu pd
2
pd =
Dynamic Replication
Dynamic Replication
Construction - Pricing options
Dynamic Replication
Dynamic Replication
Construction - Pricing options Transition probabilities of various stock price movements, pu , pd and pm . The sizes of moves up, down and middle, u, d and m=1. The payo of the option at maturity. Apply the backward algorithm derived from the risk-neutrality principle where i and j are the time and space positions respectively, so it can be calculated the price of the option at time i, Ci as the option price of an up move pu Ci+1 , plus the option price middle move by pm Ci+1 plus the option prices down move by pd Ci+1 discounted by one time step e r t , so it is possible to get the option price at any node of the tree. It is only needed to value the option at maturity and then do it backwards. fi,j = e r t [pu fi+1,j+1 + pm fi+1,j + pd fi+1,j1 ]
Dynamic Replication
Comparison of the convergence speed between a binomial and a trinomial tree
Dynamic Replication
Dynamic Replication
Comparison of the convergence speed between a binomial and a trinomial tree The main dierence between binomial and trinomial trees is just the number of transitions probabilities allowed for pricing a stock. When a small number of tree steps is used the trinomial model tends to give more accurate results than the binomial model. As the number of steps increases the results from the binomial and trinomial models (for vanilla options) rapidly converge. When a trading desk is asked for pricing a security, time (and price) are the relevant variable for making money.
Dynamic Replication
Dynamic Replication
Matlab Algorithm
Dynamic Replication
Dynamic Replication
Replication cost
Dynamic Replication
Dynamic Replication
Replication cost
Dynamic Replication

Class 8 Dynamic Replication

Загружено:

Сведения о документе

Исходное описание:

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Class 8 Dynamic Replication

Загружено:

Авторское право:

Доступные форматы

Model Description Single-Period tree Multi-period tree Calibration Extensions Trinomial Trees

Universidad Adolfo Ibanez

April 25, 2011

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Model Description Risky and risk-free asset

Fuenzalida, Lafuente, Ortiz

Model Description Risky and risk-free asset Relation between u, d, and r .

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Arrow-Debreu Securities Multi-period tree

Fuenzalida, Lafuente, Ortiz

Arrow-Debreu Securities Multi-period tree Convexity: Impossibility of static replication

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Risky and risk-free asset Relation between u, d, and r .

Fuenzalida, Lafuente, Ortiz

Risky and risk-free asset Relation between u, d, and r .

Fuenzalida, Lafuente, Ortiz

General derivative Pay-o Option European Vanilla Call

Fuenzalida, Lafuente, Ortiz

General derivative Pay-o Option European Vanilla Call No arbitrage

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Pay-o replication Condition At t1 there are two possibilities: Fu = (1 + r ) + S0 u or Fd = (1 + r ) + S0 d

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Figure: Single-PeriodReplication tree Dynamic

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz

Fuenzalida, Lafuente, Ortiz