Scribd
Upload
28 views2 pages

Lecture 12

1) The lognormal distribution is used to model stock prices over time, where the log of the price follows a normal distribution. 2) The expected value and variance of a lognormally distributed random variable X are derived. The expected value is exp(μ + σ^2/2) and the variance is exp(2μ + σ^2)(exp(σ^2) - 1). 3) These results are applied to a stock price S_t which is lognormally distributed, giving the expected value of S_t as S_0exp(μt) and its variance as S_0exp(2μt)(exp(σ^2t) - 1).

Uploaded by

Tanu Dixit
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
28 views2 pages

Lecture 12

1) The lognormal distribution is used to model stock prices over time, where the log of the price follows a normal distribution. 2) The expected value and variance of a lognormally distributed random variable X are derived. The expected value is exp(μ + σ^2/2) and the variance is exp(2μ + σ^2)(exp(σ^2) - 1). 3) These results are applied to a stock price S_t which is lognormally distributed, giving the expected value of S_t as S_0exp(μt) and its variance as S_0exp(2μt)(exp(σ^2t) - 1).

Uploaded by

Tanu Dixit
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 2

Lecture 12 A few topics on the lognormal distribution We saw in Lecture 11 that the distribution of the stock price follows

a lognormal 1 2 distribution, i.e. St = S0 exp ( )t + Bt , so Yt = ln St has a normal distribution 2 1 2 with mean ( )t and standard deviation t . We now want to derive formulae for 2 the expected value and variance of the stock price itself. Proposition 1. Let X have a lognormal distribution such that ln X : N ( , 2 ) . Then 1 E ( X ) = exp( + 2 ), V ( X ) = exp(2 + 2 )(exp( 2 ) 1) (1) 2 Proof. We begin by finding the probability density function of the variable X. Denote ( x )2 1 ( x, , ) = exp , < x < 2 2 2 ( x, , ) =

(t , , )dt

the probability density function and cumulative distribution function, respectively, for a normal random variable with mean and standard deviation . In order to derive the probability density function of X we start with its cumulative distribution function. Notice that X > 0 , so for x > 0 we have: F ( x) = P ( X x) = P (ln( X ) ln x) = (ln x, , ) (2) Now the probability density function of X can be found by differentiating in (2): dF ( x ) d (ln x, , ) 1 1 f ( x) = = = (ln x, , ) = (ln x, , ) dx dx x x 2 (ln x ) 1 1 = exp , x > 0 (3) x 2 2 2 Then (ln x ) 2 1 1 E ( X ) = xf ( x)dx = x exp dx x 2 2 2 0 0 (u ) 2 u 1 exp 2 2 2 e du 2 (u ) 2u 2 (u 2 )2 2 2 4 1 1 = du = du exp exp 2 2 2 2 2 2 2 4 2 2 2 2 + 1 (u ) = exp 2 exp 2 2 du = exp + 2 . (4) 2 2 The last equality above follows from the fact that (u 2 ) 2 1 exp du = 1 2 2 2 =

(u 2 )2 1 exp is the probability density of a normal variable with mean 2 2 2 2 + and standard deviation . In order to compute the variance we need E ( X 2 ) . We have as E ( X 2 ) = x 2 f ( x)dx = x (ln x ) 2 1 exp dx 2 2 2 0 0 (u ) 2 u 1 = eu exp e du 2 2 2 1 = 2

(u ) 2 4u 2 (u 2 2 )2 4 2 + 4 4 1 + du = du exp exp 2 2 2 2 2 2 2 2 4 2 2 4 + 4 1 (u 2 ) 2 = exp du = exp(2 + 2 ) 2 exp 2 2 2 2 Then 2 V ( X ) = E ( X 2 ) ( E ( X ) ) = exp(2 + 2 ) ( exp( 2 ) 1) (5)

(4) and (5) prove (1). QED Now applying Proposition 1 to the stock price, we obtain: S E t = exp( t ), or E ( St ) = S0 exp( t ) (6) and S0 S 2 V t = exp(2 t ) ( exp( 2t ) 1) , V ( St ) = S0 exp(2 t ) ( exp( 2 t ) 1) (7) S0

You might also like