Engineering Probability and Statistics Examples

Instructor: Dr. Zeinab Amin

Example 1

Presume that a sample of monthly sales revenues in thousands of dollars for five months
is found to be 56, 67, 52, 45, and 67. Find the
a. arithmetic mean, geometric mean and the root-mean-square value;
b. median and mode;
c. variance, standard deviation and range;
d. coefficient of variation.

Solution

1 n 56  67  52  45  67
a. Arithmetic mean X  
n i 1
Xi 
5
 57.4 thousands of dollars.
1 1
X 1 X 2  X n   X 1 X 2  X n  n   56  67  52  45  67  5  56.74
n
Geometric mean
thousands of dollars.

The root mean square value

1 n 2

n i 1
Xi 
5
56  672  522  452  672   58.04
1 2

thousands of dollars.
b. To get the median you need to arrange the data in an ascending order
45, 52, 56, 67, 67
The median is the value in the middle which is 56.
Note that in case the number of observations was even the median will be the average
of the two middle observations.
The mode is the observation which has the highest frequency, that is 67
1 n
c. The variance S 
2
  X i  X 2
 n  1 i 1

1
4
 
 56  57.4 2   67  57.4 2   52  57.4 2   45  57.4 2   67  57.4 2  92.3
1 n
The standard deviation S    X i  X 2
 n  1 i 1

1
4
 
 56  57.4 2   67  57.4 2   52  57.4 2   45  57.4 2   67  57.4 2  9.61
thousands of dollars.
The range is the difference between the largest and smallest observation 67-45=22
S 9.61
d. Coefficient of variation CV   100   100  16.74%
X 57.4

Example 2

1
Members of the Heart-to Heart Health Club are charged dues on the basis of their average
weight. Of the 60 members, 12 weighed 110 pounds, 25 weighed 120 pounds, 18
weighed 150 pounds and the rest registered 180 pounds. If the members must pay $5 for
each pound they weigh on the average, how much must each member pay?

Solution

12  110    25  120  18  150   5  180 7920

Xw    132 pounds
60 60
Each member must pay 5  132  $660 .

Example 3

Of 10 executives, 3 are to be selected to serve as a president, vice-president, and

treasurer. How many different selections are possible?

Solution
10
P3  720

Example 4

Of the 12 employees at Worldwide Travel Services, 7 have had special training. If 5

employees are to be sent to Europe, what is the probability that 3 will be among those
with the special training?

Solution
7
C3 5 C2 35  10 350
12
   0.442
C5 792 792

Example 5

If the students on the boat have three red flags, four yellow flags, and two blue flags to arrange
on a vertical pole, what is the number of possible signals?

Solution

 9  9!
    1260
3 4 2  3!4!2!

Example 6

2
In a group of 200 college students, 138 are enrolled in a course in psychology, 115 are
enrolled in a course in sociology, and 91 are enrolled in both. If a student is selected
randomly what is the probability the student is enrolled in either courses?

Solution

P: Student enrolled in psychology

S: Student enrolled in sociology

138 115 91
P P    0.69 P S    0.575 P P  S    0.455
200 200 200
P P  S   P P   P S   P P  S   0.81

Example 7

A broker knows from past experience that the probability that a client will buy stock is 65
percent. The probability that the client will buy a government bond if he or she already owns
stock is 35 percent. What is the probability that a client owns both?

Solution

S: The client will buy stock

G: The client will buy a government bond

P  S   0.65 P  G S   0.35 P S  G   ?
P  S  G   P  S   P  G S   0.65  0.35  0.2275

Example 8

Three friends, Joe, Hugh and Rae, will individually and independently attempt to solve
the same problem. The probabilities that each of them is successful are 0.9, 0.8, and 0.7
respectively.
a. What is the probability that the problem is solved by Rae alone?
b. What is the probability that the problem is solved by only one of the three?
c. What is the probability the problem will be solved by at least one of them?

Solution

J: Joe will solve the problem

H: Hugh will solve the problem
R: Rae will solve the problem

P  J   0.9 P H   0.8 P R   0.7

a. P  R  J  H   0.9  0.2  0.3  0.054
P R  J  H   P R  J  H   P R  J  H 
b.
  0.9  0.2  0.3   0.1  0.8  0.3   0.1  0.2  0.7   0.092
c. 1  P R  J  H   1   0.1  0.2  0.3  0.994

3
Example 9

In 1981, M. Radelet reported the following observed data in the American Sociological
Review (vol. 46, pp. 918-927). These data record the results of 326 murder trials held in
20 Florida counties in the years 1976 and 1977, in each of which the defendant was
convicted; his article was concerned with the imposition of the death penalty in murder
trials for white defendants versus black defendants. The data reported are given in the
following table.

Race of Death Penalty

Defendant Victim Yes No Total
White 19 132 151
White
Black 0 9 9
White 11 52 63
Black
Black 6 97 103
Total 36 290 326

a. If one trial was selected at random what is the probability that a death penalty is
imposed?
b. If one trial was selected at random what is the probability that the defendant is white?
c. If one trial was selected at random what is the probability that the victim is black?
d. If a white defendant was randomly selected what is the probability that he will receive
the death penalty?
e. If the victim is black, what is the probability that a death penalty will be imposed?

Solution

36
a.
326
160
b.
326
112
c.
326
19
d.
160
6
e.
112

Example 10

Roads A, B, and C are the only escape routes from a state prison. Prison records show
that, of the prisoners who tried to escape, 30% used road A, 50% used road B, and 20%
used road C. These records also show that 80% of those who tried to escape via A, 75%

4
of those who try to escape via B, and 92% of those who try to escape via C were
captured. What is the probability that a prisoner who was captured used road C?

Solution

A: The prisoner used road A.

B: The prisoner used road B.
C: The prisoner used road C.
R: The prisoner was captured.

P  A  0.30 P  B   0.50 P  C   0.20

P  R A  0.80 P  R B   0.75 P  R C   0.92
P C R   ?
P C  R  P C  R 
P C R   
P R  P A  R   P B  R   P C  R 
P C   P R C 

P  A  P  R A  P  B   P  R B   P  C   P  R C 
0.20  0.92 0.184
   0.23
 0.30  0.80   0.50  0.75   0.20  0.92 0.799

Example 11

A probability distribution of the claim sizes for an auto insurance policy is given in the
table below:

Claim size Probability

20 0.15
30 0.10
40 0.05
50 0.20
60 0.10
70 0.10
80 0.30
a. Find F  40  and F  55 ;
b. Find E  X  , V  X  ,  X and the coefficient of variation of claim size.

Solution

a. F  40  P  X  40  0.3

F  55  P X  55  0.5
b.

5
 Claim size Probability xp  x  x 2 p x 
x p x 
20 0.15 3 60
30 0.10 3 90
40 0.05 2 80
50 0.20 10 500
60 0.10 6 360
70 0.10 7 490
80 0.30 24 1920
Total 55 3500
E X    xp x   55

 
V  X   E X 2   E X 
2

 
E X 2   x 2 p  x   3500

V  X   3500   55  475

2

 X  V  X   3500   55 2  475  21.79

 X 21.79
Coefficient of variation=   39.6%
 55

Example 12

The time it takes for a student to finish an aptitude test (in hours) has the density function
6 x  1 2  x  1 x  2

f ( x)  
 0 elsewhere

a. Determine E  X  , V  X  ,  X and the coefficient of variation of the time it takes for a
randomly selected student to finish the aptitude test;
b. Find F  x  .

Solution

a.

6
 2 2
E X    xf  x dx   x6 x  1 2  x  dx
1 1
2 2
  6 x 3x  x 2  2 dx   18x
2
 6 x 3  12 x dx
1 1
2
 18 x3
6x 12 x4 2
 3
     
 3 4 2 1 2

2 2
E X 2    x f  x dx 
2
 x 6 x  1 2  x  dx
2

1 1
2 2
  6 x 2  3 x  x 2  2 dx   18 x
3
 6 x 4  12 x 2 dx
1 1
2
 18 x4
6x 12 x5 3

      2.3
 4 5 3 1

V  X   E  X 2    E  X    2.3  1.5  0.05

2 2

 X  V  X   0.05  0.22

 X 0.22
Coefficient of variation =   0.1467  14.67%
 1.5

b.
x x
F  x   P X  x    f  t dt   6 t  1 2  t  dt
1 1
x x
 18t 6t 3 2

  6 3t  t 2  2 dt     12t   9 x 2  2 x 3  12 x  5
1  2 3 1

Example 13

Let X 1 , X 2 , X 3 be independent random variables. E  X 1   1 , E  X 2   2 ,

E  X 3   1 , V  X 1   1 , V  X 2   3 , V  X 3   5 . Let U  X 1  2 X 2  X 3 , Find E  U  ,
V  U  and  U .

Solution

7
E  U   1  4  ( 1)  4
V U   1   4  3  1  5  18
 U  18  4.24

Example 14

An insurance company classifies each policy as High Risk or Standard. 70% of the
policies are Standard and the rest are High Risk. In a group of 5 policies selected at
random, calculate the probability that there are at least two Standard policies.

Solution

P X  2   p 2  p 3  p 4  p 5  1   p  0   p 1 
 0 5 1 4

 1  5 C 0  0.7   0.3  5C1  0.7   0.3  0.969
This answer is derived by substituting in the Binomial formula.

Example 15

The trees in a forest have a mean height of 39.5 feet with a standard deviation of 2.2 feet.
Assuming that the distribution of these trees has roughly the shape of a normal
distribution, find
a. What percentage of the trees are at least 40 feet tall?
b. What percentage of the trees are between 38 and 40 feet tall?
c. Find the height below which are the shortest 30 percent of the trees.
d. Find the height above which are the tallest 20 percent of the trees.

Solution

X: The height of the tree

  39.5
  2.2
 X   40  39.5 
a. P X  40  P    P Z  0.2   0.4207
  2.2 
b.
38  39.5 X   40  39.5 
P 38  X  40  P     P  0.7  Z  0.2   0.3213
 2.2  2.2 
c.
P  X  x   0.30 x here is unknown
 X   x  39.5  x  39.5
P    P Z  z   0.3 where z 
  2.2  2.2

From the Z table we find z  0.55

x  39.5
Therefore  0.55 hence x  38.29 feet.
2.2
d.

8
P  X  x   0.20 x here is unknown
 X   x  39.5  x  39.5
P    P  Z  z   0.2 where z 
  2.2  2.2
From the Z table we find z  0.85

x  39.5
Therefore  0.85 hence x  41.37 feet.
2.2

Example 16

A bottling machine can be regulated so that it discharges an average of 6 of ounces per

bottle. It has been observed that the amount of fill dispensed by the machine is normally
distributed with   1 ounce. A sample of 9 filled bottles is randomly selected from the
output of the machine on a given day and the ounces of fill measured for each. Find the
probability that the sample mean will exceed the population mean for that particular
setting.

Solution

X: Amount of fill dispensed by the machine

 6  1 n  9
Note that  is the population mean while X is the sample mean,  is the population
standard deviation while S is the sample standard deviation.

 X  66
P  X     P  X  6   P    P Z  0   0.5

 n 1 9 

Example 17

Suppose that a company collects data for advertising expenditures and sales revenue for 5
months, as shown in the table below

Sales Advertising
($1,000s) ($100s)
450 50
380 40
540 65
500 55
420 45

a. Compute and interpret the regression model;

b. Compute the standard error of the estimates;
c. Compute confidence intervals for the slope and for the intercept;
d. Compute the sample correlation coefficient;
e. What does this model tell you about the relationship between advertising and sales?

Solution

9
 Sales Y Advertising XY X2 Y2
X
450 50 22500 2500 202500
380 40 15200 1600 144400
540 65 35100 4225 291600
500 55 27500 3025 250000
420 45 18900 2025 176400
Y  229 X  255  XY  11920 X 2
 13375 Y 2
 1064900
0 0

a.

n5
Now use the formulas you have in your booklet on pages 3 and 4 under the title linear
regression to calculate
255
X   51
5
2290
Y   458
5
S xy  119200 
 255  2290  2410
5

S xx  13375 
 255  370
2

S yy  1064900 
 2290 
2
 16080
5
2410
bˆ   6.5135
370
aˆ  458   6.5135  51  125.8115
The regression line is then
Y  125.8115  6.5135 X
For every $100 increase in advertising, the sales will increase by 6.5135 $1000s. When
advertising is zero, the sales will be 125.8115 $1000s.

 370  16080   2410 2

b. S e2   127.477
370   5  2 
c. Confidence interval for the intercept
t , n  2
We need . We get this value from the t table on page 7 in your booklet. If
2

  0.05 , look for 0.025 in the first raw and n  2  3 in the first column. The required
value of t is 3.182. The interval is given by
 1 512 
125.8115  3.182     127.477
 5 370 

10
125.8115  96.6
 29.2115, 222.4115 

Confidence interval for the slope

The interval is given by

127.477
6.5135  3.182
370
6.5135  1.8677
 4.6458,8.3812
2410
d. R   0.988
370  16080
There is a very strong positive relationship between advertising and sales.
e. R 2  0.988 2  97.62%
Advertising explains 97.62% of the variation in sales.

Example 18

A researcher is planning to do a hypothesis test with H 0 :   0 and H1 :   0 . The

researcher would like the probability of type I error to be 0.05 and the probability of a
type II error to be 0.20. Moreover, she would like to be able to detect a difference
between the hypothesized value 0 and the true mean value 1 of 5. Assume the
population standard deviation is 17.5. Approximately, what approximate sample size is
necessary?

Solution

Probability of Type I error is   0.05

Probability of Type II error is   0.20
The difference between the hypothesized and the true value is  0  1  5
The population standard deviation is   17.5
Sample size n  ?
The formula you need is on page 4 in your booklet
 Z  2  Z   2  2 1.96  1.2816 2 17.5 2
n  =129
 1   2  2 52

Example 19

A real estate developer intends to build a large shopping mall. He may estimate the mean
family income in the area as an indicator of expected sales. A sample of 100 families
yields a mean of $35,500. Presume the population standard deviation is $7,200. Find a
95% confidence interval for the population mean income  .

Solution

11
n  100 X  35,500   7,200
You want to calculate a confidence interval for  . Note that you have two formulas (A)
and (B) at the top of page 5 in your booklet. Formula (A) is used when the population
standard deviation is known while formula (B) is used when the population standard
deviation is not known.
Here  is known so we use formula (A) with Z 2 =1.96
So the interval is
7,200
35,500  1.96 
100
 34,088.8, 36,911.2 

Example 20

The weights of 25 packages shipped through United Parcel Service had a mean of 3.7
pounds and a standard deviation of 1.2 pounds. Find the 95% confidence interval for the
mean weight of all packages. Package weights are known to be normally distributed.

Solution

n  25 X  3.7 S  1.2
You want to calculate a confidence interval for  . Note that you have two formulas (A)
and (B) at the top of page 5 in your booklet. Formula (A) is used when the population
standard deviation is known while formula (B) is used when the population standard
deviation is not known.
Here  is unknown so we use formula (B) with t 2 =2.064 (we used  2  0.025 and
n  1  24 )
So the interval is
1.2
3.7  2.064 
25
 3.205, 4.195
Example 21

Two methods for teaching reading were applied to two randomly selected groups of
elementary school children and were compared on the basis of a reading comprehension
test given at the end of the learning period. The sample means and population variances
computed from the test scores are shown in the accompanying table. Find a 95%
confidence interval for 1   2 .

Statistic Method 1 Method 2

Number of children in group 40 50
x 64 69
2 52 71

Solution

12
Notice in this problem that the population variances are known so we use formula (A) on
page 5 under confidence intervals for the difference between two means. Also notice that
in formula (A) we use the Z table
n1  40 n2  50 X 1  64 X 2  69  1  52  22  71
2

From the values of Z on page 5 we find z  1.96

We now have all what we need to substitute in formula (A)
We get   8.23,1.77  . Since the two limits are negative then 1   2  0 . The second
method is superior.

Example 22

A man who moves to a new city sees that there are two routes he could take to work. A
neighbor who has lived there a long time tells him Route A on average will be faster than
Route B. The man decides to experiment. Each day he flips a coin to determine which
way to go, driving each route 20 days. He finds that Route A takes an average of 40
minutes with standard deviation 3 minutes, and Route B takes an average of 43 minutes
with standard deviation 2 minutes. Distributions of travel times for the routes are roughly
normal and the variances are assumed equal. Construct a 99% confidence interval for
 A   B . Should the man believe the old-timer’s claim?

Solution

Notice in this problem that the population variances are not known so we use formula (B)
on page 5 under confidence intervals for the difference between two means. Also notice
that in formula (B) we use the t table
n1  20 n2  20 X 1  40 X 2  43 S1  3 S2  2
Degrees of freedom  n1  n2  2  38
From the t table on page 7 we find t38, 0.005  2.576
We now have all what we need to substitute in formula (B)
We get   5.077,0.923 . Since the two limits are negative then 1   2  0 . Therefore
the man should believe the old-timer’s claim that Route A on average will be faster than
Route B.

Example 23

Assume that the time in days required for maturation of seeds of a species of guardiola, a
flowering plant found in Mexico, is normally distributed with mean  and variance  2 .
A random sample of 13 seeds yielded a mean of 18.97 days and variance 10.7. Find a
90% confidence interval for  2 .

Solution

n  13 X  18.97 S 2  10.7
We use the formula on page 5 under confidence intervals for the variance
We get the following values from the table on page 9
 02.05,12  21.0261  02.95,12  5.22603

13
We now have all what we need to substitute in the formula
We get  6.105,24.57  .

Example 24

A particular brand of candy-coated chocolate comes in five different colors that we shall
denote as A1   brown , A2   yellow , A3   orange , A4   green and A5   coffee .
Let pi equal the probability that the color of a piece of candy selected at random
belongs to Ai , i  1, 2, ..., 5 . At significance level   0.05 test the null hypothesis
H 0 : p1  0.4, p2  0.2, p3  0.2, p4  0.1, p5  0.1
using a random sample of 580 pieces of candy whose colors yielded the respective
frequencies 224, 119, 130, 48, and 59.

Solution

Color pi Oi Ei  npi
Brown 0.4 224 232
Yellow 0.2 119 116
Orange 0.2 130 116
Green 0.1 48 58
Coffee 0.1 59 58
Total 1 580 580
 O  Ei  2
2   i  3.785
Ei
 02.05, 4  9.488 . The calculated value of chi-square < tabulated value. We do not reject
H0 .

14