Standard Normal
Normal
X
Shift
Scaling
0
1
Y
Y
Standard Normal
Normal
X
Shift
Scaling
0
1
Y
Y
Figure 2.13: Illustration of the relationship between a Normal distributed random variable and a
standard Normal distributed random variable.
The Lognormal Distribution
A random variable Y is said to be lognormal distributed if the variable Zln( )Y is Normal
distributed. It thus follows that if an uncertain phenomenon can be assumed to originate from
a multiplicative effect of several uncertain contributions then the probability distribution for
the phenomenon can be assumed to be lognormal distributed.
The lognormal distribution has the property that if:
1
i
n a
i
i
PY
. (2.33)
and all are independent lognormal random variables with parameters Yi (i, )i and 'i 0 as
given in Table 2.7 then also is lognormal with parameters: P
1
n
P ii
i
((a
(2.34)
22
1
n 2
P ii
i
) a)
(2.35)
Properties of the Expectation Operator
It is useful to note that the expectation operation possess the following properties, where
and are constants and
ab,
c X is a random variable:
12 () () 1 () 2 ()
Ec c
EcX cE X
Ea bX a bE X
Eg X g X Eg X Eg X
(2.36)
The implication of the last equation is that expectation, like differentiation or integration, is a
linear operation. This linearity property is useful since it can be used, for example, to find the
following formula for the variance of a random variable X in terms of more easily calculated
quantities: