Professional Documents
Culture Documents
Descriptive Statistics
Discrete Probability Distributions
Continuous Probability Distributions
Stochastic Processes
The relationship for determining the Mean or Expected Value is the same as the
Dr. A. Ben Hamza
Concordia University
f(x)
0
1
Uniform distribution
(e.g. soil texture)
0
Exponential distribution
(e.g. event rainfall)
0.3
f (x)
A Discrete PDF
0.2
Probability that x = 2
0.25
0.15
0.1
0
1
2
3
4
x
Discrete distribution
(e.g. number of severe storms)
Only discrete
values (integers)
are possible
Var ( X ) E ( X 2 ) ( E ( X )) 2 2
Smallest values
are most likely
e
f ( x)
x!
x 0,1, 2,...
Where the parameter >0 is the mean number of successes in the interval.
The mean and variance of the Poisson distribution are
and
( x ) 2
2 2
x
2
We can use the following function to convert any normal random variable to a
standard normal random variable
= 1.5?
0
f (0) P( X 0)
e x e 1.51.50
0.2231
x!
0!
Some advice: always
draw a picture!
11
f ( x)
f(x)
for x
10
12
If
X ~ LN(,),
(0.76) 0.776373
(1.3) ?
then
Y= ln (X) ~ N(,)
( 3) 1 (3) ?
(3.86) ?
ln x
F(x) P( X x) F
where F(z) is the cumulative probability distribution function of N(0,1)
13
Lognormal Distribution
Mean or Expected Value of X
1
2
2
X E( X ) e
Median of X
X E( X ) e
2 2 2
e 1
X e
2
2
e 5.005
149.16
median e
Standard Deviation of X
15
X e 2 (e 1)
1
2
223
14.933
14
P ( X 120) 1 P ( X 120)
ln 120 5.0
)
0. 1
1 F (2.13)
1 P( Z
1 0.0166
0.9834
16
17
ln 110 5.0
ln 130 5.0
P (110 X 130) P (
Z
)
0 .1
0 .1
E (Y ) np 10 0.983
9.83
P( 2.99 Z 1.32)
F (1.32) F ( 2.99)
0.0934 0.0014
0.092
(d)
ln xms 5.0
) 0.05 ,
0.1
,
P ( Z 1.64) 0.05
and
P (Z
so that
ln xms 5.0
1.64
0.1
xms 125.964
therefore
p P ( X 120)
1 P ( X 120)
ln 120 5.0
)
0.1
1 F ( 2.12)
1 P(Z
19
18
20
Exponential Distribution
A random variable X is defined to be exponential random variable (or
say X is exponentially distributed) with positive parameter if its
probability density function is given by:
e x
f ( x)
0
Note:
if x 0, 0
if x 0
f ( x) dx e x dx e x
F ( x) P( X x) f (t ) dt 0
1 0.0170
For
x 0, F ( x) 0 dt 0
0.983
For
x 0, F ( x) et dt et 1 e x
1 e x
F (x)
0
if x 0
if x 0
21
Exponential Distribution
Expectation:
E[ X ] xf ( x )dx
x e x dx
E[ X ] xe x
( e x ) dx
0
e x
1
1
x
e
( )
1
dx e x
f ( x; , )
0
( e x )2 x dx 2 xe dx
x e dx
2 1
2
2
Var [ X ] E [ X 2 ] ( E[ X ]) 2
2 1
1
2
2
22
x 0,
x 2 e x dx x 2 de x
Integration by part:
E[ X 2 ] x 2e x
for
Otherwise
Variance:
E [ X 2 ] x 2 f ( x) dx
Gamma Distribution, if
x de x
Integration by part:
23
Gamma Distribution
24
Gamma Function
Definition
For 0
hours
( )
P (10 X 15)
F (15) F (10)
P(10 X 15)
e (0.05)(10) e (0.05)(15)
e 0.5 e 0.75
0.1341
1 x
e dx
For any
1, ( ) ( 1) ( 1)
1
2
n, ( n) ( n 1)!
25
1
2, 0.5
0.8
0.6
1, 1
0.4
2, 2
2, 1
0.2
0
0
x
2
If X~G(, ), then
Mean or Expected Value: E ( X )
Standard Deviation:
26
27
28
In general, the value X t might depend on the quantity X t-1 at time t-1,
or even the value X s for other times s < t.
,..., X t k
x1 ,..., x k P X t
x1 ,..., X t k x k
29
31
X1
X4
X3
X2
4 th Event
Occurs
time
1
t=0
S2
Xi
S3
i 1
Xi
S4
i 1
i 1
identical distribution
Xn depicts the time elapsed between the (n-1)th event and nth event occurrences
Sn depicts a random variable for the time at which the nth event occurs
Define N(t) as the number of events that have occurred up to some arbitrary time t.
Xi
i 1
Cov X t , X s E X t t X s s
S1
Cov X t , X s
t2 s2
The counting process { N(t), t>0 } is called a Poisson process if the interoccurrence times X1, X2, follow the exponential distribution
30
Counting process
32
Sunday
N (t=10 min) = 2
1st Bus
Arrival
X1=5 min
2nd Bus
Arrival
X2=4 min
3 rd Bus
Arrival
X3=7 min
4 th Bus
Arrival
X4=2 min
time
t=0
S1 = 5 min
S2 = 9 min
S3 = 16 min
S 4 = 18 min
33
35
Monday
1 st Bus
Arrival
N (t=10 min) =4
2 nd Bus
Arrival
3rd Bus
Arrival
X3=4 min
5th Bus
Arrival
4th Bus
Arrival
X4=2 min
X5=6 min
time
t=0
S1 = 1 min S2 = 3 min
S 3 = 7 min
S4 = 9 min
S 5 = 15 min
34
Tuesday
N (t=10 min) =1
1st Bus
Arrival
X1=10 min
2nd Bus
Arrival
X2=6 min
time
t=0
S 1 = 10 min
S 2 = 16 min