Вы находитесь на странице: 1из 24

Probability Distributions Used in

Simulation
Michael A. Carchidi
January 29, 2009

Binomial Distribution With Parameters n and p: Models the number of


successes in n Bernoulli trials, when the trials are independent with common
success probability p.
Probability Mass Function, Range Space, Application: The pmf of this distribution is given by
P (x; n, p) =

n x
p (1 p)nx
x

with a range space of


RX = {x | x = 0, 1, 2, . . . , n}.
For example, the number of defective computer chips found in a lot of n
identical chips, given that the probability that any single chip is defective is
p.
The Parameters: The parameter p equals the probability of success per
trial and can take on any real value between 0 and 1, inclusive while the
parameter n equals the number of independent trials and can take on any
positive-integer value.
Mean and Variance: The mean and variance are given by
E(X) = np
respectively.

&

V (X) = np(1 p),

Cumulative Density Function: The cdf is given by


F (x; n, p) =

x
X

k=0

n k
p (1 p)nk .
k

Example Plot Of The pmf : An example plot of the pmf using n = 10 and
p = 0.4 is presented below.
25
0.2
15
0.1
05

0 0

10

Plot of P (x; n, p) versus x for n = 10 and p = 0.4

Geometric Distribution With Parameter p: Models the number of Bernoulli


trials required to achieve the first success.
Probability Mass Function, Range Space, Application: The pmf of this distribution is given by
P (x; p) = (1 p)x1 p
with a range space of
RX = {x | x = 1, 2, 3, . . .}.
For example, the number of identical computer chips that must be inspected
to find the first defective chip, given that the probability that any single chip
is defective is p.
The Parameters: The parameter p equals the probability of success per trial
and can take on any real value between 0 and 1, inclusive.
2

Mean and Variance: The mean and variance are given by


E(X) =

1
p

&

V (X) =

1p
,
p2

respectively.
Cumulative Density Function: The cdf is given by
F (x; p) =

x
X

k=0

p(1 p)k1 = 1 (1 p)x

Example Plot Of The pmf and cdf : An example plots of the pmf and cdf
using p = 0.4 are presented below.
0.4

0.3

0.2

0.1

10

Plot of P (x; p) versus x for p = 0.4


1
0.9
0.8
0.7
0.6
0.5
0.4
2

10

Plot of F (x; p) versus x for p = 0.4


3

The Memoryless Property: The geometric distribution is the only discrete


distribution having the memoryless property, which states that
Pr(X > s + t | X > s) = Pr(X > t)
for all s 0 and t 0.

Negative Binomial (Pascal) Distribution With Parameters r and p: Models the number of Bernoulli trials required to achieve the first k successes.
Probability Mass Function, Range Space, Application: The pmf of this distribution is given by

x1 r
P (x; r, p) =
p (1 p)xr
r1
with a range space of
RX = {x | x = r, r + 1, r + 2, . . .}.
For example, the number of identical computer chips that must be inspected
to find the first r defective chips, given that the probability that any single
chip is defective is p.
The Parameters: The parameter p equals the probability of success per
trial and can take on any real value between 0 and 1, inclusive while the
parameter r equals the number of successes and can take on any integer
value larger than zero.
Mean and Variance: The mean and variance are given by
E(X) =

r
p

&

V (X) =

r(1 p)
,
p2

respectively.
Cumulative Density Function: The cdf is given by
F (x; r, p) =

x
X

k=r

k1 r
p (1 p)kr .
r1

Example Plot Of The pmf : An example plot of the pmf using r = 5 and
p = 0.4 is presented below.

0.1
08
06
04
02

0 0

10

15

20

25

Plot of P (x; r, p) versus x for r = 5 and p = 0.4

Hypergeometric Distribution With Parameters g, b and n: Models the


number of good items when choosing n items (without replacement) from a lot of
g + b items, g of which are good and b of which are bad.
Probability Mass Function, Range Space, Application: The pmf of this distribution is given by

P (x; g, b, n) =

with a range space of

g
x

b
nx

g+b
n

RX = {x | x = xmin , xmin + 1, xmin + 2, . . . , xmax }


where
xmin = max(n b, 0)

and

xmax = min(g, n).

For example, the number of good computer chips out of n inspected, in a


lot of g good and b bad chips.
The Parameters: The parameter g equals the number of good objects in
the lot while the parameter b equals the number of bad objects and the
parameters n equals the number of items chosen from the lot.
5

Mean and Variance: The mean and variance are given by

g
E(X) = n
g+b

&

g
V (X) = n
g+b

b
g+b

g+bn
.
g+b1

respectively.
Cumulative Density Function: The cdf is given by
F (x; g, b, n) =


g
k

x
X

k=xmin

b
nk

g+b
n

Example Plot Of The pmf : An example plot of the pmf using g = 10, b = 15
and n = 9 is presented below.

0.3
25
0.2
15
0.1
05
0 0

Plot of P (x; g, b, n) versus x for g = 10, b = 15 and n = 9

Poisson Distribution With Parameter : Models the number of independent


events that occur in a fixed amount of time or space.
Probability Mass Function, Range Space, Application: The pmf of this distribution is given by
e x
P (x; ) =
x!
with a range space of
RX = {x | x = 0, 1, 2, 3, . . .}.
6

For example, the number of customers that arrive to a store during the first
t time units, or the number of defects found in 30 square meters of sheet
metal.
The Parameters: The parameter equals the average number of independent events occurring in the fixed time or space interval and can take on
any real value larger than zero.
Mean and Variance: The mean and variance are given by
&

E(X) =

V (X) = ,

respectively.
Cumulative Density Function: The cdf is given by
F (x; ) =

x
X
e k

k=0

k!

Example Plots Of The pmf : An example plot of the pmf using = 2 is


presented below.

25
0.2
15
0.1
05

0 0

Plot of P (x; ) versus x for = 2


Discrete Uniform Distribution With Parameter m: Models m events, all
with the same probability of occurring.
Probability Mass Function, Range Space, Application: The pmf of this distribution is given by
1
P (x; m) =
m
with a range space of
RX = {x | x = 1, 2, 3, . . . .m}.
The Parameters: The parameter m equals the number of events and can
take on any integer value larger than zero.
Mean and Variance: The mean and variance are given by
E(X) =

m+1
2

&

V (X) =

m2 1
,
12

respectively.
Cumulative Density Function: The cdf is given by

F (x; m) =

for
for
for
for
..
.

0,
1,
2,
3,
..
.

x<1
1x<2
2x<3
3x<4
..
.

m 1, for m 1 x < m
m,
for m x

Example Plots Of The pmf : An example plot of the pmf using = 2 is

presented below.
0.5
0.4
0.3
0.2
0.1

0 0

Plot of P (x; m) versus x for m = 5

Continuous Uniform Distribution With Parameters a and b: Models a


random draw between a and b.
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by
f (x; a, b) =
with a range space of

1
U[a, b)
ba

RX = {x | a x < b}.
The Parameters: The parameter a equals the smallest possible value while
b equals the largest possible value.
Mean and Variance: The mean and variance are given by
E(X) =

a+b
2

&

V (X) =

respectively.
Cumulative Density Function: The cdf is given by
F (x; a, b) =
9

xa
.
ba

(b a)2
,
12

Example Plots Of The pdf and cdf : Example plots of the pdf and cdf using
a = 2 and b = 5 are presented below.
0.5
0.4
0.3
0.2
0.1

0 0

Plot of f (x; a, b) versus x for a = 2 and b = 5

1
0.8
0.6
0.4
0.2

0 0

Plot of F (x; a, b) versus x for a = 2 and b = 5

Normal (Guassian) Distribution With Parameters and 2 : Models the


distribution of a process that can be though of as the sum of a number of independent component processes. For example, the time to assemble a product which is
the sum of the times required for each assembly operation.
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by
(

1
1 x
f (x; , 2 ) = exp
2

2
10

2 )

N(, 2 )

with a range space of


RX = {x | < x < +}.
The Parameters: The parameter (for any < < +) equals the
mean of the distribution while 2 (for any 0 < 2 ) equals the variance.
Mean and Variance: The mean and variance are given by
V (X) = 2 ,

&

E(X) =
respectively.

Cumulative Density Function: The cdf is given by


(

1 Zx
1 t
exp
F (x; , ) =
2

2
2

2 )

x
dt

where

1 Z x 1 t2
(x) =
e 2 dt
2
and these are tabulated in Table A.3 (pages 574 and 575) on the text.

Example Plots Of The pdf and cdf : Example plots of the pdf and cdf using
= 2 and 2 = 4 are presented below.
0.2

0.15

0.1

0.05

-4

-2

0 0

Plot of f (x; , 2 ) versus x for = 2 and 2 = 4

11

1
0.8
0.6
0.4
0.2

-4

-2

0 0

Plot of F (x; , 2 ) versus x for = 2 and 2 = 4


The Central Limit Theorem: Let X1 , X2 , X3 , . . ., Xn be a sequence of
independent random variables with E(Xi ) = i and V (Xi ) = i2 for i =
1, 2, 3, . . . , n. In addition let
X = X1 + X2 + X3 + + Xn .
Then in the limit of large n, X N(, 2 ) with
=

n
X

2 =

and

i=1

n
X

i2 .

i=1

Lognormal Distribution With Parameters and 2 : Models the distribution of a process that can be though of as the product of a number of independent
component processes. For example, the rate of return on an investment, when
interest is compounded, is the product of the returns for a number of periods.
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by

1
1
exp
f (x; , 2 ) =
2
x 2

ln(x)

with a range space of

RX = {x | 0 < x < +}.


12

!2

The Parameters: The parameters (for any < < +) and 2 (for
any 0 < 2 ) are related to the mean and variance of the distribution as
shown below.
Mean and Variance: The mean and variance are given by
E(X) = e+

2 /2

V (X) = (e 1)e2+ ,

&

respectively.
Cumulative Density Function: The cdf is given by
1
F (x; , 2 ) =
2

Z x
0

which can be written as

1
1
exp
t
2

!2
ln(t)

dt

ln(x)
.
F (x; , ) =

Example Plots Of The pdf : An example plot of the pdf using = 3 and
2 = 4 is presented below.

07
06
05
04
03
0

Plot of f (x; , 2 ) versus x for = 3 and 2 = 4


The Central Limit Theorem: Let Y1 , Y2 , Y3 , . . ., Yn be a sequence of independent random variables with E(Yi ) = i and V (Yi ) = i2 for i = 1, 2, 3, . . . , n.
In addition let
Y = Y1 + Y2 + Y3 + + Yn
13

Then in the limit of large n, Y Y (, 2 ) with


=

n
X

2 =

and

i=1

n
X

i2 ,

i=1

and X = eY , with
X = X1 X2 X3 Xn = eY1 eY2 eY3 eYn = eY1 +Y2 +Y3 ++Yn .
is lognormal with parameters and 2 .

Exponential Distribution With Parameter : Models the time between


independent events, or a process time which is memoryless (knowing how much
time has passed gives no information about how much additional time will pass
before the process is complete).
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by
f (x; ) = ex
with a range space of
RX = {x | 0 < x < +}.
For example, the time between the arrival of a large number of customers
who act independently of each other.
The Parameters: The parameter , which can take on any positive real
value, is the reciprocal of the mean and standard deviation of the distribution as shown below.
Mean and Variance: The mean and variance are given by
E(X) =

&

V (X) =

respectively.
Cumulative Density Function: The cdf is given by
F (x; ) = 1 ex .
14

1
,
2

Example Plots Of The pdf and cdf : Example plots of the pdf and cdf using
= 1 are presented below.
1
0.8
0.6
0.4
0.2

0.5

1.5

2.5

3.5

Plot of f (x; ) versus x for = 1


1
0.8
0.6
0.4
0.2

0 0

0.5

1.5

2.5

3.5

Plot of F (x; ) versus x for = 1


The Memoryless Property: The exponential distribution is the only continuous distribution having the memoryless property, which states that
Pr(X > s + t | X > s) = Pr(X > t)
for all s 0 and t 0.
The Relationship Between The Exponential and Poisson Distributions: If
time between events is exponentially distributed, then the number of events
in a fixed period of time is Poisson.

15


Gamma Distribution With Parameters and : An extremely flexible distribution used to model nonnegative random variables.
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by
() 1 x
x e
()
with a range space of

where

f (x; , ) =

() =

t1 et dt

RX = {x | 0 < x < +}.


The Parameters: The parameter and can take on any positive real
values.
Mean and Variance: The mean and variance are given by
1
1
E(X) =
&
V (X) = 2 ,

respectively.
Cumulative Density Function: The cdf is given by

() Z x 1 t
F (x; , ) =
t e
dt.
() 0

Example Plots Of The pdf : Example plots of the pdf using = 1 and = 1,
2, and 3 are presented below.
1
0.8
0.6
0.4
0.2

0 0

0.5

1.5

2.5

3.5

Plots of f (x; , ) versus x for = 1 and = 1, 2, and 3


16

A Relationship Between The Exponential and Gamma Distributions: When


is an integer, the gamma distribution is related to the exponential distribution in the following way. If the random variable X is the sum of
independent, exponentially distributed random variables, each with parameter , then X has a gamma distribution with parameters and .

Beta Distribution With Parameters , , a and b: An extremely flexible distribution used to model bounded (fixed upper and lower limits) random
variables.
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by
f (x; , ; a, b) =

( + + 1)!(x a) (b x)
!!(b a)++1

with a range space of


RX = {x | a x b}.
The Parameters: The parameters a and b can take on any real values as
long as a < b. In addition, a is the minimum value of X while b is the
maximum possible value of X. The parameters and can take on any
positive real values.
Mean and Variance: The mean and variance are given by
E(X) =

( + 1)a + ( + 1)b
++2

&

V (X) =

( + 1)( + 1)(b a)2


,
( + + 2)2 ( + + 3)

respectively.
Cumulative Density Function: The cdf is given by

( + + 1)! Z x (t a) (b t)
( + + 1)!
xa
F (x; , ; a, b) =

dt
++1
!!
(b a)
!!
ba
a

where
(z; , ) =

Z z
0

17

u (1 u) du

Example Plots Of The pdf and cdf : An example plot of the pdf using a = 1,
b = 3 and, = 2 and = 3 is presented below.
1
0.8
0.6
0.4
0.2

1.5

2.5

3.5

Plots of f (x; , ; a, b) versus x for


a = 1, b = 3, = 2 and = 3

Erlang Distribution With Parameters k and : Models processes that can


be viewed as the sum of several independent exponentially distributed processes.
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by
f (x; k, ) =
with a range space of

(k)k k1 kx
x e
(k 1)!

RX = {x | 0 < x}.

For example, a computer network fails when a computer and two backup
computers fail, and each has a time for failure that is exponentially distributed.
The Parameters: The parameter can take on any real positive values while
k is restricted to just the positive integers.
Mean and Variance: The mean and variance are given by
1
1
E(X) =
&
V (X) = 2 ,

k
respectively.
18

Cumulative Density Function: The cdf is given by


k1
X ekx (kx)i
(k)k Z x k1 kt
F (x; k, ) =
t e
dt = 1
(k 1)! 0
i!
i=0

Example Plots Of The pdf and cdf : Example plots of the pdf using = 1
and = 1, 2, and 3 are presented below.
1
0.8
0.6
0.4
0.2

0 0

0.5

1.5

2.5

3.5

Plots of f (x; k, ) versus x for = 1 and k = 1, 2, and 3


A Relationship Between The Exponential and Erlang Distributions: If the
random variable X is the sum of k independent, exponentially distributed
random variables, each with parameter k, then X has an Erlang distribution with parameters k and .

Weibull Distribution With Parameters , , and : Models the time to


failure for components. For example, the time to failure for a disk drive.
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by

x
f (x; , , ) =

x
exp

with a range space of


RX = {x | < x}.
19

For example, a computer network fails when a computer and two backup
computers fail, and each has a time for failure that is exponentially distributed.
The Parameters: The parameters , , and can take on any positive real
values.
Mean and Variance: The mean and variance are given by

1
E(X) = + 1 +

2
1+

&

V (X) =

1
1+

!)

respectively.
Cumulative Density Function: The cdf is given by zero for x < , and
(

x
F (x; , , ) = 1 exp

for x
Example Plots Of The pdf and cdf : Example plots of the pdf and cdf using
= 1, = 3 and = 2.

1
0.8
0.6
0.4
0.2

0 2

2.5

3.5

4.5

Plot of f (x; , , ) versus x for = 1, = 3 and = 2

20

1
0.8
0.6
0.4
0.2

0 2

2.5

3.5

4.5

Plot of F (x; , , ) versus x for = 1, = 3 and = 2

Triangular Distribution With Parameters a, b, and c: Models a process


when only the minimum, most likely and maximum values of the distribution are
known. For example, the minimum, most likely, and maximum time required to
test a product.
Probability Density Function, Range Space, Application: The pdf of this
distribution is given by
2(xa)

(ba)(ca) ,

f (x; a, b, c) =

with a range space of

2(cx)
,
(cb)(ca)

for a x b
for b x c

RX = {x | a x c}.
The Parameters: The parameters a, b, and c can take on any real values
as long as a < b < c, and these represent the minimum, most likely and
maximum values of the distribution, respectively.
Mean and Variance: The mean and variance are given by
E(X) =

a+b+c
3

&

V (X) =

respectively.
21

(a2 + b2 + c2 ) (ab + bc + ac)


,
18

Cumulative Density Function: The cdf is given as

0,

(xa)2

(ba)(ca) ,

F (x; a, b, c) =

for

xa

for a x b

(cx)2
,
(cb)(ca)

for b x c
for

1,

cx

Example Plots Of The pdf : An example plot of the pdf using a = 5, b = 15


and c = 40.

05
04
03
02
01

0 0

10

20

30

40

Plot of f (x; a, b, c) versus x using


a = 5, b = 15 and c = 40

Trapezoidal Distribution With Parameters a, b, c and d: Generalizes the


Uniform and Triangular Distributions. A continuous random variable X has a
trapezoidal distribution with parameters a b c d if its shape is a trapezoid
connecting the points
(a, 0) (b, h) (c, h) (d, 0).
Probability Density Function, Range Space, Application: The pdf of this
22

distribution is given by

h(x a)/(b a), for a x b


for b x c .

f (x; a, b, c, d) = h,

with

h(d x)/(d c), for c x d

h=
and this has a range space of

2
(c + d) (a + b)

RX = {x | a x d}.
The Parameters: The parameters a, b, c, and d can take on any real values
as long as a < b < c < d. The parameters a and d represent the minimum
and maximum values of the distribution, respectively.
Mean: The mean is given by
1
E(X) =
3

(c2 + cd + d2 ) (a2 + ab + b2 )
.
(c + d) (a + b)

Cumulative Density Function: The cdf is given as

0,

1 (xa)2

h ba ,

for x a
for a x b

F (x; a, b, c, d) = 12 h(2x a b), for b x c .

1 12 h (dx)
,
dc

for c x d

1,

for d x

23

Example Plots Of The pdf : An example plot of the pdf using a = 1, b = 2,


c = 4 and d = 7.
25
0.2
15
0.1
05

0 0

Plot of f (x; a, b, c, d) versus x using


a = 1, b = 2, c = 4, d = 7

24

Вам также может понравиться