Advanced Business Statistics
MGSC-372
MGSC 372
The Lognormal Distribution
The Lognormal Distribution
A continuous random variable X follows a lognormal distribution if its natural logarithm, ln(X), follows a normal distribution. The lognormal distribution is an asymmetric distribution with interesting applications for modeling the probability distributions of stock and other asset prices
The Lognormal Distribution
• Properties of the lognormal distribution
– Skewed to the right
– Strictly positive (i.e. bounded below by 0)
– The lognormal distribution may be used to model data on asset prices (note that prices are bounded below by 0)
Histograms of lognormal variable X and normal distribution ln(X)
The Lognormal Distribution
• The lognormal distribution is described by two parameters, its mean and variance, as in the case of a normal distribution
• The mean of a lognormal distribution X is given by
E ( X ) e
1 2
2
• The variance of a lognormal distribution X is given by
Var (X ) e
(2 2 )
2
(e 1)
where and 2 are the mean and variance of the normal distribution of the ln(X) variable and e 2.718 is the natural base for logarithms.
• The median of a lognormal distribution X is given by eµ .
The Lognormal Distribution
Recall that the exponential and logarithmic functions are inverse functions so that
ln( x) y e y x
This relationship is used to switch between the lognormal variable X and the normal variable y = ln X
Lognormal Distribution - Example
The material failure mechanism for a specific metal alloy has determined that the ductile strength X of the material has a lognormal distribution with parameters = 5 and = 0.1
(a) Compute E(X) and Var(X)
(b) Compute P(X > 120)
(c) Compute P(110 X 130)
(d) What is the value of median ductile strength?
(e) If the smallest 5% of strength values were unacceptable, what would the minimum acceptable strength be?
Lognormal Distribution –Example Solution
(a)
2 u 2 e5.005 e5.005 149.16
E ( X ) e
2
2
2
u
Var ( X ) e
(e
1) 223
(b)
P( X 120) 1 P( X 120) ln120 5.0
1 P( Z
)
0 .1
1 F ( 2.13)
1 0.0166
0.9834
Lognormal Distribution –Example Solution
(c) P (110 X 130) P ( ln 110 5.0 Z ln 130 5.0 )
0.1
P ( 2.99 Z 1.32)
F ( 1.32) F ( 2.99)
0.0934 0.0014
0.092
(d)
Median eu e5 148.41
0.1
Lognormal Distribution –Example Solution
(e) The value of X, call it L,, for which P( X L) 0.05 is determined as follows: ln L 5.0
P( Z
) 0.05
0.1
P( Z 1.64) 0.05 ln L 5.0
1.64
0.1
L 125.964
Example: Relative Asset Prices and the Lognormal Distribution
• Consider the relative price of an asset between periods 0 and 1, defined as S1/S0, which is equal to 1 + R0,1
• For example, if S0 = $30 and S1 = $34.5, then the relative price is
$34.5/$30 = 1.15 or 1+ 0.15, meaning that the holding period return R0,1 is 15%
• The continuously compounded return rt,t+1 associated with a holding period return of Rt,t+1 is given by the natural log of the relative price
rT ,T 1
ST 1
ln
ln 1 RT ,T 1
ST
Example: Relative Asset Prices and the Lognormal Distribution
•
For the above example, the continuously compounded return is given by r0,1 = ln($34.5/$30) = ln(1.15) = 0.1397 or 13.97%, which is lower than the holding period return of 15%.
•
To generalize, note that between periods 0 and T, r0,T = ln(ST/S0) or we can write r ST S0 e 0,T
•
Note that
ST
ST ST 1 ST 2
S2 S1
.
.
...... .
S0 ST 1 ST 2 ST 3
S1 S0
Example: Relative Asset Prices and the Lognormal Distribution
• Recall that ln(XY) = ln(X) + ln(Y) so that
S
S
S
S ln T ln T ln T 1 ln 1
ST 1
ST 2
S0
S0 or, equivalently,
r0 ,T rT 1,T rT 2 ,T 1 r0 ,1
We see that the mean continuously compounded return between periods 0 and T is the sum of the continuously compounded returns of