PROBABILITY
The above results may be extended. For example, if the random variables
Xi,i=1, 2 ,...,n, are distributed asXi∼N(μi,σ^2 i) then the random variable
Z=
∑
iciXi(where theciare constants) is distributed asZ∼N(
∑
iciμi,
∑
ic
2
iσ
2
i).
30.9.2 The log-normal distribution
If the random variableXfollows a Gaussian distribution then the variable
Y=eXis described by alog-normaldistribution. Clearly, ifXcan take values
in the range−∞to∞,thenYwill lie between 0 and∞. The probability density
function forYis found using the result (30.58). It is
g(y)=f(x(y))
∣
∣
∣
∣
dx
dy
∣
∣
∣
∣=
1
σ
√
2 π
1
y
exp
[
−
(lny−μ)^2
2 σ^2
]
.
We note that μandσ^2 are not the mean and variance of the log-normal
distribution, but rather the parameters of the corresponding Gaussian distribution
forX. The mean and variance ofY, however, can be found straightforwardly
using the MGF ofX, which readsMX(t)=E[etX]=exp(μt+^12 σ^2 t^2 ). Thus, the
mean ofYis given by
E[Y]=E[eX]=MX(1) = exp(μ+^12 σ^2 ),
and the variance ofYreads
V[Y]=E[Y^2 ]−(E[Y])^2 =E[e^2 X]−(E[eX])^2
=MX(2)−[MX(1)]^2 =exp(2μ+σ^2 )[exp(σ^2 )−1].
In figure 30.15, we plot some examples of the log-normal distribution for various
values of the parametersμandσ^2.
30.9.3 The exponential and gamma distributions
The exponential distribution with positive parameterλis given by
f(x)=
{
λe−λx forx> 0 ,
0forx≤ 0
(30.116)
and satisfies
∫∞
−∞f(x)dx= 1 as required. The exponential distribution occurs nat-
urally if we consider the distribution of the length of intervals between successive
events in a Poisson process or, equivalently, the distribution of the interval (i.e.
the waiting time) before the first event. If the average number of events per unit
interval isλthen on average there areλxevents in intervalx, so that from the
Poisson distribution the probability that there will be no events in this interval is
given by
Pr(no events in intervalx)=e−λx.