This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Chapter 5 ArbitrageFree Pricing in OnePeriod Finite Models The simplest financial models involving random evolution of prices are those in which there are only two trading times and the prices of the basic securities are modelled as random variables on a finite probability space. Although such models are too simplistic to accurately model realworld financial markets, they will allow us to introduce some important ideas (such as riskneutral pricing) in the simplest possible setting. Moreover, these very basic models can be used as building blocks in more sophisticated models. 5.1 OnePeriod Binomial Model In the oneperiod binomial model there are two times, t = 0 and t = 1, a bank, and a single stock that pays no dividends. We can borrow or invest any amount of money at the bank between t = 0 and t = 1 at the oneperiod interest rate r 0. Capital in the bank account evolves according to the rule B 1 = B (1 + r ) , where B is the initial capital; B > 0 corresponds to an investment, whereas B < indicates a loan. The initial price per share of the stock is S . It is assumed that we can buy or sell short as many shares of the stock as we please at time 0 at the price S per share. The price S 1 of the stock at time 1 is not known at time 0. We assume that there are two possible values of S 1 . Consequently it is appropriate to model S 1 as a random variable on a probability space ( , P ) in which contains two elements. We think of these two elementary events as head and tail resulting from the toss of a coin. An appropriate sample space is therefore = { H,T } . We do not assume that the coin is fair, i.e. we do not assume that P ( H ) = P ( T ). We do, however, assume that P ( H ) > , P ( T ) > . 131 We also assume that S 1 ( H ) negationslash = S 1 ( T ); otherwise the stock is just another bank account (and the noarbitrage principle would tell us that the interest rate for the stock must also be r .) We refer to P as the reference probability measure , or the actual probability measure , in order to distinguish it from another probability measure P , called a pricing measure (or riskneutral measure ), that will be introduced later to compute arbitragefree prices. This model is extremely simple. It is certainly a stretch to believe that the price of the stock at the terminal time can take only two possible values, and that these values are known at the initial time. However, it is very useful to study (and completely understand) the oneperiod binomial model. Indeed, many of the important features of this model have analogues in more sophisticated models. In particular, oneperiod binomial models can be strung together to form multiperiod binomial models, which are somewhat more realistic. A continuoustime model can be obtained taking a suitable limit of multiperiod binomial models. The key to analyzing multiperiod models, is the complete analysis of oneperiod models....
View
Full
Document
This note was uploaded on 11/08/2011 for the course 21 270 taught by Professor Schaffer during the Spring '10 term at Carnegie Mellon.
 Spring '10
 SCHAFFER

Click to edit the document details