Lecture-- 5

Lecture
Discrete Probability Distributions
Random Variables
Lecture-- 5
Lecture 
 Discrete Probability Distributions
Discrete Probability Distributions  Expected Value and Variance
 Binomial Probability Distribution
 Poisson Probability Distribution
Course Instructor: Rumana Hossain
 Hypergeometric Probability Distribution
Deparment of Physical Sciences; SECS
.40

.30

.20

.10

0 1 2 3 4

Slide 1 Slide 2

Random Variables Example: JSL Appliances

 A random variable is a numerical description of the
outcome of an experiment.
 A random variable can be classified as being either
discrete or continuous depending on the numerical
values it assumes.
 A discrete random variable may assume either a
finite number of values or an infinite sequence of
values.
 A continuous random variable may assume any
numerical value in an interval or collection of
intervals.
 Discrete random variable with a finite number of
values
Let x = number of TV sets sold at the store in one day
where x can take on 5 values (0, 1, 2, 3, 4)
 Discrete random variable with an infinite sequence of
values
Let x = number of customers arriving in one day
where x can take on the values 0, 1, 2, . . .
We can count the customers arriving, but there is no
finite upper limit on the number that might arrive.

Slide 3 Slide 4

1
 Discrete Probability Distributions Example: JSL Appliances

 The probability distribution for a random variable  Using past data on TV sales (below left), a tabular
describes how probabilities are distributed over the representation of the probability distribution for TV
values of the random variable. sales (below right) was developed.
 The probability distribution is defined by a
probability function,
function, denoted by f(x), which provides Number
the probability for each value of the random variable. Units Sold of Days x f(x)
 The required conditions for a discrete probability 0 80 0 .40
function are: 1 50 1 .25
f(x) > 0 2 40 2 .20
f(x) = 1 3 10 3 .05
 We can describe a discrete probability distribution 4 20 4 .10
with a table, graph, or equation. 200 1.00

Slide 5 Slide 6

Example: JSL Appliances Discrete Uniform Probability Distribution

 Graphical Representation of the Probability  The discrete uniform probability distribution is the
Distribution simplest example of a discrete probability
distribution given by a formula.
.50  The discrete uniform probability function is
Probability

.40
f(x) = 1/n
1/n
.30 where:
.20 n = the number of values the random
variable may assume
.10
 Note that the values of the random variable are
equally likely.
0 1 2 3 4
Values of Random Variable x (TV sales)

Slide 7 Slide 8

2
 Expected Value and Variance Example: JSL Appliances

 The expected value,

value, or mean, of a random variable is  Expected Value of a Discrete Random Variable
a measure of its central location.
• Expected value of a discrete random variable: x f(x) xf((x)
xf
0 .40 .00
E(x) =  = xf
xf((x)
1 .25 .25
 The variance summarizes the variability in the values
2 .20 .40
of a random variable.
• Variance of a discrete random variable: 3 .05 .15
4 .10 .40
Var(xx) =  2 = (x - )2f(x)
Var(
E(x) = 1.20
 The standard deviation,
deviation, , is defined as the positive
square root of the variance. The expected number of TV sets sold in a day is 1.2

Slide 9 Slide 10

Example: JSL Appliances Binomial Probability Distribution

 Variance and Standard Deviation  Properties of a Binomial Experiment

of a Discrete Random Variable • The experiment consists of a sequence of n
identical trials.
x x- (x - )2 f(x) (x - )2f(x)
• Two outcomes, success and failure
failure,, are possible on
0 -1.2 1.44 .40 .576 each trial.
1 -0.2 0.04 .25 .010 • The probability of a success, denoted by p, does
2 0.8 0.64 .20 .128 not change from trial to trial.
3 1.8 3.24 .05 .162 • The trials are independent.
4 2.8 7.84 .10 .784
1.660 =  
The variance of daily sales is 1.66 TV sets squared
squared..
The standard deviation of sales is 1.2884 TV sets.

Slide 11 Slide 12

3
 Example: Evans Electronics Binomial Probability Distribution

 Binomial Probability Distribution  Binomial Probability Function

Evans is concerned about a low retention rate for n!
employees. On the basis of past experience, f ( x)  p x (1  p) ( n  x )
x !( n  x )!
management has seen a turnover of 10% of the
hourly employees annually. Thus, for any hourly where:
employees chosen at random, management estimates f(x) = the probability of x successes in n trials
a probability of 0.1 that the person will not be with n = the number of trials
the company next year.
p = the probability of success on any one trial
Choosing 3 hourly employees a random, what is
the probability that 1 of them will leave the company
this year?
Let::
Let p = .10, n = 3, x = 1

Slide 13 Slide 14

Example: Evans Electronics Example: Evans Electronics

 Using the Binomial Probability Function  Using the Tables of Binomial Probabilities
n! p
f ( x)  p x (1  p) ( n  x ) n x .10 .15 .20 .25 .30 .35 .40 .45 .50
x !( n  x )!
3 0 .7290 .6141 .5120 .4219 .3430 .2746 .2160 .1664 .1250
3! 1 .2430 .3251 .3840 .4219 .4410 .4436 .4320 .4084 .3750
f (1)  ( 0.1)1 ( 0. 9 ) 2 2 .0270 .0574 .0960 .1406 .1890 .2389 .2880 .3341 .3750
1!( 3  1)!
3 .0010 .0034 .0080 .0156 .0270 .0429 .0640 .0911 .1250
= (3)(0.1)(0.81)

= .243

Slide 15 Slide 16

4
 Example: Evans Electronics Binomial Probability Distribution

 Using a Tree Diagram  Expected Value

First Second Third Value
Worker Worker Worker of x Probab. E(x) =  = np
L (.1) 3 .0010
Leaves (.1)  Variance
S (.9) 2 .0090
Leaves (.1) Var((x) =  2 = np
Var (1 - p)
np(1
L (.1) 2 .0090
Stays (.9)  Standard Deviation
S (.9) 1 .0810
SD( x )    np (1  p )
L (.1) 2 .0090
Leaves (.1)
S (.9) 1 .0810
Stays (.9) L (.1) 1 .0810
Stays (.9)
S (.9) 0 .7290
Slide 17 Slide 18

Geometric distribution
Example: Evans Electronics

 Binomial Probability Distribution  The geometric random variable x is defined as the

• Expected Value number of trials needed to obtain the first success.
E(x) =  = 3(.1) = .3 employees out of 3 Therefore, if success is obtained in the nth trial, this
means that the previous (n(n – 1) trials ended up in
• Variance failure.
Var(x) =  2 = 3(.1)(.9) = .27
• Standard Deviation
SD( x)    3(.1)(.9)  .52 employees

Slide 19 Slide 20

5
 Geometric distribution Geometric distribution

 If probability of success is considered to be p, the  Mean, m = E[x] = 1

probability of failure would be (1 – p). For sake of p q
simplicity, the probability of failure is often denoted  Variance, s2 = E[x2] – E[x]2 =
by q. Therefore the geometric probability distribution p2
can be written as
p(x = n) = (1 – p)n – 1p = qn – 1p
Or simply q
f(x) = qx – 1p  Standard deviation, s =
p2

Slide 21 Slide 22

Example
 The requested probability is
 The probability that a wafer contains a large particle  p(X = 125) = (0.985)1240.015 = 0.0023
of contamination is 0.015. If it is assumed that the  Average number of wafers those have to be tested before
wafers are independent, what is the probability that a contamination is found is .
exactly 125 wafers need to be analyzed before a large
particle is detected?
1 1
  67
Let X denote the number of samples analyzed until a p 0 . 015
large particle is detected. Then X is a geometric  Standard deviation of the number of wafer those have to
random variable with p = 0.01. be tested is

q 0 .985
  66
p2 0 .015 2
Slide 23 Slide 24

6
 Negative binomial distribution
Negative binomial distribution

 The negative binomial distribution is a distribution
that can be thought of as ‘reverse’ of the binomial
distribution. In binomial distribution, the number of
trials, n, is fixed, and the number of successes, x, is a
variable. We essentially look for the probability of x
successes in n trials.
 In negative binomial distribution, the number of
successes is fixed, and we look for the number of
trials that would be needed to obtain the success.
Therefore, the number of success, r, is kept fixed; and
the number trials, x, is made a variable.
 Therefore the characteristics of the problem are
 The experiment consists of a series of independent
and identical trial, each with a probability of success
p;
 The trials are observed until exactly r successes are
obtained, where r is fixed by the experimenter;
 The random variable x is the number of trials needed
to obtain r successes.

Slide 25 Slide 26

Negative binomial distribution Negative binomial distribution

Several important observations can be here.  The mean, m =

 The experiment would always end with a success; r
 There would be exactly r – 1 successes in the previous x –
1 trials, in any combination;  The variance, s2 =
p
 There would be exactly x – r failures in x – 1 trials.
 Therefore the probability distribution function can be rq
written as
p2
 x  1
f ( x )    1  p x  r p r

 r 1

Slide 27 Slide 28

7
 Example

 Cotton linters used in the production of solid rocket
propellants are subjected to a nitration process that
enables the cotton fibers to go into the solution. The
process is 90% effective in that the materials
produced can be shaped as desired in a later
processing stage with probability 0.9. What is the
probability that exactly 20 lots will be produced in
order to obtain the third defective lot?
 Here ‘success’ is obtaining a defective lot; and hence
p = 0.1 and r = 3. The probability that x = 20 is given
by
19 
f (20)   (0.9)17 (0.1)3  0.0285
2
 The expected value is 3/0.1 = 30, meaning that on an
average 30 trails would be required to produce the
third defective lot.

Slide 29 Slide 30

Poisson Probability Distribution Poisson Probability Distribution

 Properties of a Poisson Experiment  Poisson Probability Function

• The probability of an occurrence is the same for  xe
any two intervals of equal length. f ( x) 
x!
• The occurrence or nonoccurrence in any interval is
independent of the occurrence or nonoccurrence where:
in any other interval. f(x) = probability of x occurrences in an interval
 = mean number of occurrences in an interval
e = 2.71828

Slide 31 Slide 32

8
 Example: Mercy Hospital Example: Mercy Hospital

 Using the Poisson Probability Function  Using the Tables of Poisson Probabilities
Patients arrive at the emergency room of Mercy

Hospital at the average rate of 6 per hour on x 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.0
weekend evenings. What is the probability of 4 0 .1225 .1108 .1003 .0907 .0821 .0743 .0672 .0608 .0550 .0498
arrivals in 30 minutes on a weekend evening? 1 .2572 .2438 .2306 .2177 .2052 .1931 .1815 .1703 .1596 .1494
2 .2700 .2681 .2652 .2613 .2565 .2510 .2450 .2384 .2314 .2240
 = 6/hour = 3/half-
3/half-hour, x = 4 3 .1890 .1966 .2033 .2090 .2138 .2176 .2205 .2225 .2237 .2240
4 .0992 .1082 .1169 .1254 .1336 .1414 .1488 .1557 .1622 .1680
34 ( 2 . 71828 ) 3
f (4)   .1680 5 .0417 .0476 .0538 .0602 ..0668 .0735 .0804 .0872 .0940 .1008
4! 6 .0146 .0174 .0206 .0241 .0278 .0319 .0362 .0407 .0455 .0504
7 .0044 .0055 .0068 .0083 .0099 .0118 .0139 .0163 .0188 .0216
8 .0011 .0015 .0019 .0025 .0031 .0038 .0047 .0057 .0068 .0081

Slide 33 Slide 34

Hypergeometric Probability Distribution Hypergeometric Probability Distribution

 The hypergeometric distribution is closely related to  Hypergeometric Probability Function

the binomial distribution.
 With the hypergeometric distribution, the trials are  r  N  r 
  
not independent, and the probability of success x n  x 
changes from trial to trial. f ( x)    for 0 < x < r
N 
 
n
where: f(x) = probability of x successes in n trials
n = number of trials
N = number of elements in the population
r = number of elements in the population
labeled success

Slide 35 Slide 36

9
 Example: Neveready Example: Neveready

 Hypergeometric Probability Distribution  Hypergeometric Probability Distribution

Bob Neveready has removed two dead batteries
from a flashlight and inadvertently mingled them  r  N  r   2  2   2!  2! 
        
with the two good batteries he intended as x n  x   2  0   2!0! 0!2! 1
f ( x)        .167
replacements. The four batteries look identical. N 4  4!  6
     
Bob now randomly selects two of the four n 2  2!2!
batteries. What is the probability he selects the two where:
good batteries? x = 2 = number of good batteries selected
n = 2 = number of batteries selected
N = 4 = number of batteries in total
r = 2 = number of good batteries in total

Slide 37 Slide 38

End of lecture-
lecture- 5

Slide 39

10