MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

Lecture 2a: Dimension analysis of the Black-Scholes formula, No-arbitrage theory and Put-call parity
1. Dimension analysis of the Black-Scholes formula Dimension analysis is often used in physics to study the similarity structure of a certain phenomenon. For example, the air ow outside of an airplane is often controlled by a single factor called Mach number, which measures the relative velocity of the airplane over the speed of sound. The dimension analysis is introduced here in quantitative nance to study the intrinsic nature of option price and to determine the key factors that govern the price law of an option. After some analysis, we will

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

see that the non-dimensional option price is indeed controlled by only two factors, instead of six in the original Black-Scholes formula. One is the moneyness and the other one is the standard deviation of stock return over the period from now until maturity date. 1.1 Dimension Dimension is dened to be unit here. Dimensionless variable is a variable that has no dimension. We often use a square bracket [ ] to denote the dimension of a variable. For example, the dimension of your height is meter. This

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

statement can be written as [h] = m, where h is a variable height and m is meter. Similarly we can write [St] = $, [ct] = $, [K ] = $, [Ker(T t)] = $,

where $ stands for dollar. We also have [T ] = [t] = year, [r ] = 1 , year [ 2] = 1 , year [ ] = 1 , year

so that r(T t) and T t are dimensionless. The ratio between the current stock price, St, and the current

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

value of strike price, Ker(T t), is dimensionless, i.e., St Ker(T t) has no dimension. The concept of dimension has many applications. One important application is that it can be used to judge whether a complicated formula is correct or not. For example, ct = N (d1) Ker(T t)N (d2) is a wrong formula, because the rst term has no dimension and the dimension of the second term is dollar. Two variables with dierent dimensions cannot be subtracted each other.

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

The following formula ct = StN (d1) Ker N (d2) is not correct either, because the interest rate r has a dimension 1/year, er does not make sense. The independent variable of the exponential function has to be dimensionless. With a clear concept of dimension in mind, one could build an intuition with complex mathematical formula more quickly and have a less chance to make a mistake. Dimension analysis of nancial quantities helps you to remember the formula. 1.2 Moneyness We now introduce a new concept called Moneyness, which is

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

dened to be variable by following = St Ker(T t) 1. (1)

We explain why this term is called moneyness. For a call option, if St > Ker(T t), the call option is called in-the-money. The owner of the option is likely to gain if an in-the-money option is to be exercised on maturity date. In practice, the interest rate eect is small, i.e., er(T t) often uses St 1 K to dene the moneyness. But the denition (1) is more reason1

1 1, one

r(T t)

When interest rate and time to maturity are small, for example, r = 0.05, T t = e0.004 = 0.996 1, here we apply the Taylor expansion to have ex 1 + x.

1/12, then r(T t) = 0.004, and

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

able and rigorous. The dierence between current stock price and the current value of strike price St Ker(T t) is often called the intrinsic value of the call. If St = Ker(T t), the call option is at-the-money. If St < Ker(T t), the call option is out-of-the-money. In nancial markets, the most frequently traded options, e.g., the S&P 500 (SPX) options traded in Chicago Board Options Exchange (CBOE), are near the money, i.e., St Ker(T t), = St Ker(T t) 1 0.

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

The moneyness is usually small. Its value is between 0.1 and 0.1. In fact, a call option is often called at-the-money if (0.02, 0.02), in-the-money if (0.02, 0.05), deep-in-themoney if (0.05, 0.07), out-of-the-money if (0.05, 0.02), deep-out-of-the-money if (0.07, 0.05). Very few options with reasonable trading volume have an absolute moneyness larger than 0.07. With the moneyness , we rewrite d1 and d2 as follows ln(1 + ) 1 + T t, d1 = T t 2 ln(1 + ) 1 T t. d2 = T t 2 (2) (3)

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

The Black-Scholes formula is re-written in a non-dimensional form c t = (1 + )N (d1 ) N (d2 ), where Ker(T t) is the relative price of the European call option over the current value of strike price, d1 and d2 are given by (2) and (3). Please pay attention that d1 and d2 here are functions of two variables, the moneyness , the standard deviation of stock return over the period [t, T ], T t. Therefore the relative price of the option is also a function of and T t. c t = ct (4)

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

10

In the original Black-Scholes formula, the option price is a function of six variables/parameters, St, t, K, T, , r, i.e., ct = ct(St, t; K, T, , r), where a semi-column is used to separate independent variables and parameters. Variables St and t are called independent variables and they are changing with time. Variables K and T are called parameters and they have the same value for a particular contract and dier only across dierent contracts. In fact, they are used to label the contracts. Variable is a volatility parameter that measures the variability of the underlying price. Variable r is a riskfree rate parameter. It is the same for all

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

11

options on dierent underlying stocks. In the non-dimensional Black-Scholes formula (4), the relative option price is a function of two variables, the moneyness, , and the standard deviation, T t, i.e., c t = c t , T t .

By introducing the dimensionless variables, the Black-Scholes formula has been simplied from six variables to two variables due to the intrinsic nature of the option price uncovered by the mathematical tool of dimension analysis. One may observe that the interest rate eect disappear in the non-dimensional Black-Scholes formula. As a matter of

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

12

fact, it is embedded into the current value of strike price in the moneyness and the relative price of the option. 2. No-arbitrage Theory The rst fundamental principle in derivative pricing: noarbitrage. Axiom: There is no arbitrage in nancial markets. Based on this principle, we have following theorem: Theorem 1 If two portfolios have the same payo on a certain date, T , in the future (maturity date for option), they must have the same value at any time, t, before the future date, T .

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

13

Suppose the payos of portfolio 1 and 2 are P O1T and P O2T respectively and P O1T = P O2T , we need to prove that P O1t = P O2t. Proof. We use the method of proof by contradiction. Suppose P O1t > P O2t, one would be able to construct a new portfolio by taking a long position on portfolio 2 and a short position on portfolio 1, and earns a prot of P O1t P O2t at current time. On maturity date, he/she would use P O2 to delivery the obligation of the short position of P O1 and make it even since P O1T = P O2T . The nal result would be that he/she earns a prot of P O1t P O2t at current time without taking any risk.

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

14

There is an arbitrage. This contradicts to our rst principle. Therefore P O1t P O2t. Similarly, if P O1t < P O2t, One would be able to construct a new portfolio by taking a long position on portfolio 1 and a short position on portfolio 2, and generate an arbitrage P O2t P O1t at current time. Once again this contradicts to the rst principle. Therefore, P O1t = P O2t has to be true. 3. Put-call parity Put-call parity describes the relation between the price of a European call and the price of a European put.

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

15

The payo of a European call is given by cT = max(ST K, 0), while the payo of a European put is given by pT = max(K ST , 0). (6) (5)

Proposition. A portfolio of long a call and short a put with the same strike price, K , and the same maturity date, T , have the same payo as that of a long position on a forward contract with a delivery price, K , on delivery date, T , i.e., cT pT = ST K. (7)

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

16

This can be proved either graphically or algebraically. Graphic Proof. The payo diagrams of the call and put are shown in the gure below, where the horizontal axis is ST and the vertical axis is the payo of options, the strike price K is 2. By subtracting the second diagram from the rst one, one would obtain the third diagram, which is the payo of a forward contract.

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

17

3 2 1 -1 -2 -3 1 2 3 4 5

3 2 1 -1 -2 -3
3 2 1 -1 -2 -3 1 2 3 4 5

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

18

Algebraic Proof.

cT pT = max(ST K, 0) max(K ST , 0) = max(ST K, 0) + min(ST K, 0) = ST K. Consider two portfolios: Portfolio 1: long a call and short a put. Portfolio 2: long a forward contract with a delivery price K .
2

Q.E.D.

In this algebraic proof, we need the knowledge of max/min function as follows: max(a, b) = min(a, b) Example: max(1, 2) = 2 = min(1, 2) max(a, b) + min(a, b) = a + b

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

19

The current value of the portfolio 2 is St Ker(T t). From Theorem 1, we have following result Theorem 2 The current price, ct, of a European call option and the current price, pt of a European put option with the same strike price, K , and maturity, T , satises the following relationship ct pt = St Ker(T t), which is often called put-call parity. With the parity relationship, the price of a European put can (8)

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

20

be derived from that of a European call given by the BlackScholes formula, i.e., pt = ct St + Ker(T t) = StN (d1) Ker(T t)N (d2) St + Ker(T t) = St(N (d1) 1) Ker(T t)(N (d2) 1) = Ker(T t)N (d2) StN (d1),
3

(9)

where following property of the cumulative normal distribution function

N (x) + N (x) = 1 has been used in the derivation.

A rigorous proof of the property is available in the Appendix.

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

21

Theorem 3 If the price of a European call is given by ct = StN (d1) Ker(T t)N (d2), where d1 and d2 are given by
1 2 t ln S K + (r + 2 )(T t) , d1 = T t 1 2 t ln S + ( r K 2 )(T t) , d2 = T t

(10)

then the price of a corresponding European put is given by pt = Ker(T t)N (d2) StN (d1). (11)

The Black-Scholes option pricing formula (10) will be proved later in this course.

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

22

Appendix. A useful property of the cumulative normal distribution function Proposition. The cumulative normal distribution function N (x) satises the following relation: N (x) + N (x) = 1. Proof. By denition,
x

(12)

N (x) =

n(y )dy,

and n(x) is an even function, i.e., n(x) = n(x),

MFIN 7003 Mathematical Techniques of Finance I: Jin E Zhang

23

therefore
x

N (x) + N (x) = = =

n(y )dy +
x

x x

n(y )dy,
x x

Let y = z in the second term n(y )dy n(y )dy = 1

n(z )dz =

n(y )dy +

n(z )dz

Q.E.D.

Applying the formula at x = 0 gives N (0) + N (0) = 1, 1 N (0) = . 2