Академический Документы
Профессиональный Документы
Культура Документы
Basic Problem:
To relate the magnitude of extreme events to
their frequency of occurrence through the use
of probability distributions.
FREQUENCY ANALYSIS
Basic Assumptions:
(a) Data analyzed are to be statistically independent
& identically distributed - selection of data (Time
dependence, time scale, mechanisms).
(b) Change over time due to man-made (eg.
urbanization) or natural processes do not alter the
frequency relation - temporal trend in data
(stationarity).
FREQUENCY ANALYSIS
Practical Problems:
(a) Selection of reasonable & simple distribution.
(b) Estimation of parameters in distribution.
(c) Assessment of risk with reasonable accuracy.
f ( x) dx 1
p( x) 1
all x
xo
f ( x) dx
all xi xo
p ( xi )
Product-Moments
rth-order product-moment of X about any reference point X=xo is defined, for continuous case,
as
r
r
r
E X xo = x xo f x x dx = x xo dFx x
= xk xo p x xk
k=1
r = 1 C r,i xi 'r i
i= 0
'r = C r,i xi r i
i= 0
x f x dx = x dF x = 1 F x
x
dx
E ak X k = ak k
k=1
k=1
X
=
k k
k=1
k=1
Mode (xmo) - the value of a RV at which its PDF is peaked. The mode, xmo, can be obtained by
f x
solving
=0
x xmo
x
(a) Uni-modal distribution
fx(x)
x
(b) Bi-modal distribution
2
x
=
2
f xx dx
2
Var
k=1
X k = ak2 k2
k=1
Skewness Coefficient
Measures asymmetry of the PDF of a random variable
Skewness coefficient, x, defined as
3 E X
x = 1. 5 =
x3
2
x x mo
x
xmo xmd x
xxmo=xmd
fx(x)
(c) Negatively skewed, x<0
x
xmd xmo
Measure of
Central
Location
Definition
Mean, Expected value
E(X)=x
Variance, Var(X)=2= x2
Continuous Variable
x f ( x) dx
x
(x x ) fx (x) dx
2
x
Second
Dispersion
Third
Asymmetry
3
( x x ) f x ( x) dx
x = 3 / x3
4
(x )
x
Fourth
Peakedness
Kurtosis, x
Excess coefficient, x
x = 4 / x4
xx
x xk p( xk )
x xi / n
all x's
2
2
xk x Px xk
x all x ' s
x = xx
Skewness coefficient, x
Sample Estimator
x Var ( X )
Standard deviation, x
x Var ( X )
Coefficient of variation,x x = xx
Skewness
Discrete Variable
f x ( x ) dx
1
xi x
n 1
Cv s x
x = 3 / x3
g = m3 / s3
4 ( xk x ) 4 px ( xk ) m4
x = 4 / x4
xx
m3
all x's
1
xi x
n 1
3 ( xk x )3 px ( xk )
all x's
s2
n
xi x
(n 1)(n 2)
n (n 1)
xi x
(n 1)(n 2)(n 3)
k = m4 / s4
Kurtosis (x)
Measure of the peakedness of a distribution.
Related to the 4th central product-moment as
4 E X
x = 2 =
x4
2
x2 1 x
f N x | x , x2 =
1
2 x
1
exp 2
x x
Standardized Variable:
Z
(z ) 0.3
0.2
0.1
0
-3
-2
-1
(a) x = 1.0
x=0.
3
x=0.
6
0
x=1.3
3
x
4
0.7
x =1.65
0.6
(b) x = 1.30
0.5
0.4
x =2.25
fLN(x)
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1 ln( x) -
1
ln x
=
exp -
,x>0
ln x
2 ln x x
2
0.3
x =4.50
0.2
0.1
0
3
x
fLN(x)
x
F EV 1 x | , = exp exp
x
= 1 exp exp +
for maxima
f EV 1 x | , =
for minima
x
x
1
exp
exp
1
exp +
exp
fEV1(y)
0.4
Max
Min
0.3
0.2
0.1
0
-4
-3
-2
-1
x
+
for maxima
for minima
- x /
0.30
0.25
0.20
fG(x) 0.15
0.10
0.05
0.00
0
10
12
14
ln x
- ln( x ) /