Slides - Lecture9
Slides - Lecture9
Sandra Nolte [email protected] KE 515 Oce Hours: TBA School of Management, University of Leicester
Outline
2/
Non-Arbitrage Theorem
We have 3 assets and two states of nature. Assume that the states of nature are: a good economy which is state 1 and a bad economy which is state 2. 1. Asset 1 (riskless): bond with price now of 1 and returns of 1+r in both states - where r is the risk free rate. 2. Asset 2 (risky): a stock with costs now S(t) and will cost in the future, at time t + 1, either S1 (t + 1) or S2 (t + 1) 3. a call on the underlying stock which costs now c(t) and will cost in the future, at time t + 1, either c1 (t + 1) or c2 (t + 1)
3/
Non-Arbitrage Theorem
The non-arbitrage theorem says that if we nd positive constants,1 , 2 so that the system of equations: 1 = S(t) c(t) = = (1 + r)1 + (1 + r)2 S1 (t + 1)1 + S2 (t + 1)2 c1 (t + 1)1 + c2 (t + 1)2
is satised, then there are no arbitrage possibilities. And if there are no arbitrage opportunities then positive constants, 1 , 2 satisfying the above system of equations can be found. Note that we can set: (1 + r)1 = p1 , where p1 is some probability and (1 + r)2 = p2 , is also some probability. Both probabilities add up to unity: p1 + p2 = 1.
4/
Non-Arbitrage Theorem If we multiply S(t) with unity 1 = 1+r 1+r 1 S(t) = 1+r 1 S(t) = 1+r Because (1 + r)1 = p1 = 1+r S(t) 1+r
1+r 1+r :
(S1 (t + 1)1 + S2 (t + 1)2 ) (S1 (t + 1)1 (1 + r) + S2 (t + 1)(1 + r)2 ) (p1 S1 (t + 1) + p2 S2 (t + 1)) and since (1 + r)2 = p2
But p1 S1 (t + 1) + p2 S2 (t + 1) is the expected value of the stock price, E[S(t + 1)] The above equation can be rewritten as: 1 E[S(t + 1)] S(t) = 1+r We can discount the risky stock value at the risk free rate, using the probabilities p1 and p2
5/
Non-Arbitrage Theorem Let us consider the following example 4 1.1 1.1 100 = 100 150 1 2 call price 0 50 Assume that we observe a call price in the market of 25. How can we check whether 25 is the non-arbitrage price for the call? Solve the system of equations! Let us write the system of equations: 1 = 100 = call price = 1.11 + 1.12 1001 + 1502 502
6/
Non-Arbitrage Theorem
From the third equation: for the call price of 25 2 = 0.5 From the second equation we get 1 = 0.25. From the rst equation you get: 1.1 0.25 + 1.1 0.5 = 0.825! 25 must not be the arbitrage free price.
7/
Non-Arbitrage Theorem
All concepts we have see so far are based on the assumption of no arbitrage Put-Call Parity, volatility smile, binomial tree,.. Under the assumption of no-arbitrage, we can write the Black-Scholes portfolio like d = rdt But arbitrage opportunities exist in the real world!
8/
Models dealing with arbitrage option pricing are the models by Ilinski & Stepanenko (1999) or Fedotov & Panayides (2005), among others. The idea here is to introduce an arbitrage return, say x, so that the return on the Black-Scholes portfolio, , becomes: d = rdt + xdt; where r is the risk free rate of interest. x can follow a specic random process like the Ornstein-Uhlenbeck process as in Ilinski & Stepanenko (1999) Named after Leonard Ornstein and George Eugene Uhlenbeck. The process is stationary, Gaussian, and Markov. Over time, the process tends to drift towards its long-term mean.
9/
This process is dened there as: dx = x(t) + (t) dt where (t) is noise and is the parameter which reects the speed the market reacts on the arbitrage opportunity. The derivative price obtained in the Ilinski models is an average derivative price
10/
Arbitrage Option Pricing Model Fedotov & Panayides (2005) assume that the uctuations around the risk free rate of interest are not restricted to follow a particular process, like the Ornstein-Uhlenbeck process. More general approach. 2 sources of uncertainty: 1. uctuation from the stock (usual geometric Brownian motion) dS = Sdt + SdW, where W is a Wiener process 2. uncertainty from a random arbitrage return from the bond dB = rBdt + (t)Bdt; where r is the risk free interest rate, B the bond price and (t) is some random process (which does NOT have to obey any specic stochastic dierential equation).
30th March 2012 11/
They assume that the variations on the risk free return, which mimic arbitrage returns, are on a dierent time scale. The time scale is on a scale of hours and denoted by As the authors remark, this time lies in between the time scale for the stock return (innitely Brownian motion uctuations) and the maturity time of the derivative (months): 0 << << T The option pricing portfolio they use is given by V = B+ SV S where V is the option price.
12/
Arbitrage Option Pricing Model Using Itos Lemma, they obtain the classical Black-Scholes equation: V V 1 2V 2 2 + rS + S = rV 2 t S 2 S But using both uncertainties they obtain: V 1 2 V 2 2 V +rS + S +r+(t)+(t) t S 2 S 2 V SV S + = rV
Solving the equation, the authors nd that the option price, when taking into account arbitrage returns will be of the general form: VBS (S) 2 U (S, )
where is dened as T , U (S, ) is a function of (t) and VBS is the Black-Scholes option price.
Other kind of Options Employee Stock Options: call options on a companys stock granted by the company to its employees, see chapter 157 in Hull. Options on Currencies: large trades are possible, strike price, expiry date and other features tailored to meet the needs of corporate treasurers Future Options: is the right not the obligation to enter into a future contract at a specied time in the future, at a specied future price , see chapter 17 in Hull. Exotic Options: Packages, Nonstandard American Options (Bermuda), Gap Options, Forward Start Options, Cliquet Options, Compound Options, Chooser Options, Barrier Options,Binary Options, Lookback options, Shout Options, Asian Options, see chapter 25 in Hull. Real Options: Options on real assets
30th March 2012 14/
Readings
Hull J. C. (2000); Options, Futures and Other Derivatives; Prentice Hall, Chapter 19. Fedotov S., & S. Panayides (2005): Stochastic Arbitrage return and its implication for option pricing, Physica A, 345, 207-217. Ilinski K. & A. Stepanenko (1999): Derivative pricing with virtual arbitrage, Quantitative Finance Papers Nr. 9902046
15/