Geoscience Reference
In-Depth Information
A.7.2.3 Exponential Distribution
e
a
x
1
R
þ
x
X
Exponential
ðaÞ$
f
X
x
ðÞ
¼
a
ðÞ:
For X with an exponential distribution,
¼
a
1
EX
:
A.7.2.4 Pareto Distribution
ðÞ
¼½
ba
b
=ð
x
bþ
1
X
Pareto
ða; bÞ$
f
X
x
Þ
1
½
a;þ/Þ
x
ðÞ
for
a
[
0e
b
[
0
:
For X with a Pareto distribution,
EX
ðÞ
¼
ab=ðb
1
Þ;
for
b
[
1
;
and
E
ðÞ
¼
þ/;
for
b
1
:
X
Pareto
ða; b; Þ$
ln
ð
X
=aÞ
Exponential
ðbÞ:
A.7.2.5 Normal Distribution
X has a normal distribution with location parameter
ʼ
and dispersion parameter
˃
ðÞ
¼
ð
p
Þ
1
2
2
r
1
2
ð
X
Normal
ðl; r
ÞÞ $
f
X
x
2
p
exp½
ðð
1
=
2
x
lÞ
=r
:
EX
¼
l:
2
Var X
ðÞ
¼
r
:
The normal distribution has the following useful properties:
2
X
Normal
ðl; r
Þ$ð
X
lÞ=r
Normal 0
ð
;
;
1
any linear combination of random variables with a normal distribution has a normal
distribution;
if X
n
is the sum of n random variables independent and identically distributed
with a distribution with expected value
ʼ
and standard deviation
˃
then the
p
Þ
distribution of
.
As a particular instance of this last property, for the summands with Bernoulli
distributions, we have
ð
X
n
n
lÞ=ðr
approaches a Normal(0,1) distribution as n
!∞
p
nq 1
X
Binomial n
ð
;
q
Þ!
ð
X
nq
Þ=
½
ð
q
Þ
Normal 0
ðÞ
;
1
Search WWH ::
Custom Search