This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Continuous Time Finance Notes, Spring 2004 – Section 2, Jan. 28, 2004 Notes by Robert V. Kohn, Courant Institute of Mathematical Sciences. For use in connec tion with the NYU course Continuous Time Finance. In Section 1 we discussed how Girsanov’s theorem and the martingale representation the orem tell us how to price and hedge options. This short section makes that discussion concrete by applying it to (a) options on a stock which pays dividends, and (b) options on foreign currency. We close with a brief discussion of Siegel’s paradox. For topics (a) and (b) see also Baxter and Rennie’s sections 4.1 and 4.2 – which are parallel to my discussion, but different enough to be well worth reading and comparing to what’s here. For a discussion of Siegel’s paradox and some related topics, see chapter 1 of the delightful book Puzzles of Finance: Six Practical Problems and their Remarkable Solutions by Mark Kritzman (J. Wiley & Sons, 2000, available as an inexpensive paperback). ******** Options on a stock with dividend yield. You probably already know from Derivative Securities how to price an option on a stock with continuous dividend yield q . If the stock is lognormal with volatility σ and the riskfree rate is (constant) r then the “riskneutral process” is dS = ( r q ) S dt + σS dw , and the time0 value of an option with payoff f ( S T ) and maturity T is e rT E RN [ f ( S T )]. From the SDE we get dE RN [ S ] /dt = ( r q ) E RN [ S ], so E RN [ S ]( T ) = e ( r q ) T S . This is the forward price , i.e. the unique choice of k such that a forward with strike k and maturity T has initial value 0. (Proof: apply the pricing formula to f ( S T ) = S T k .) When the option is a call, we get an explicit valuation formula using the fact that if X is lognormal with mean E [ X ] = F and volatility s (defined as the standard deviation of log X ), then E [( X K ) + ] = F N ( d 1 ) KN ( d 2 ) (1) with d 1 = ln( F/K ) + s 2 / 2 s , d 2 = ln( F/K ) s 2 / 2 s . There is of course a similar formula for a call (easily deduced by putcall parity). What does it mean that the “riskneutral process is dS = ( r q ) S dt + σS dw ?” What happens when the dividends are paid at discrete times? We can clarify these points by using the framework of Section 1. Remember the main points: (a) there is a unique “equivalent martingale measure” Q , obtained by an application of Girsanov’s theorem; and (b) this Q is characterized by the property that the value V t of any tradeable asset satisfies V t /B t = E [ V T /B T F t ] where B t is the value of a riskfree moneymarket account, i.e. it solves dB = rB dt with B (0) = 1. Put differently: V/B is a Qmartingale....
View
Full
Document
This note was uploaded on 01/02/2012 for the course FINANCE 347 taught by Professor Bayou during the Fall '11 term at NYU.
 Fall '11
 Bayou
 Finance

Click to edit the document details