Sta 2200 Notes

Time is precious, but we do not know yet how precious it really is.
We will only know when we are no longer able to take advantage of it…
STA 2200 PROBABILITY AND STATISTICS II

Purpose By the end of the course the student should be able to solve problems involving
probability distributions of a discrete or a continuous random variable.
Objectives
By the end of this course the student should be able to;
(1) Define the probability mass, density and distribution functions, and to use these to
determine expectation, variance, percentiles and mode for a given distribution
(2) Appreciate the form of the probability mass functions for the binomial, geometric,
hyper-geometric and Poisson distributions, and the probability density functions for the
uniform, exponential, gamma , beta and normal, functions, and their applications
(3) Apply the moment generating function and transformation of variable techniques
(4) Apply the principles of statistical inference for one sample problems.
DESCRIPTION
Random variables: discrete and continuous, probability mass, density and distribution
functions, expectation, variance, percentiles and mode. Moments and moment generating
function. Moment generating function and transformation Change of variable technique for
univariate distribution. Probability distributions: hyper-geometric, binomial, Poisson, uniform,
normal, beta and gamma. Statistical inference including one sample normal and t tests.
Pre-Requisites: STA 2100 Probability and Statistics I, SMA 2104 Mathematics for Science
Course Text Books
1) RV Hogg, JW McKean & AT Craig Introduction to Mathematical Statistics, 6th ed.,
Prentice Hall, 2003 ISBN 0-13-177698-3
2) J Crawshaw & J Chambers A Concise Course in A-Level statistics, with worked examples,
3rd ed. Stanley Thornes, 1994 ISBN 0-534- 42362-0
Course Journals:
1) Journal of Applied Statistics (J. Appl. Stat.) [0266-4763; 1360-0532]
2) Statistics (Statistics) [0233-1888]
Further Reference Text Books And Journals:

a) HJ Larson Introduction to Probability Theory and Statistical Inference(Probability and
Mathematical Statistics) 3rd ed., Wiley, 1982
b) Uppal, S. M. , Odhiambo, R. O. & Humphreys, H. M. Introduction to Probability and
Statistics. JKUAT Press, 2005
c) I Miller & M Miller John E Freund’s Mathematical Statistics with Applications, 7th ed.,
Pearsons Education, Prentice Hall, New Jersey, 2003 ISBN: 0131246461
d) Statistical Science (Stat. Sci.) [0883-4237]
e) Journal of Mathematical Sciences
f) The Annals of Applied Probability
Proverbs 21:5 The plans of the diligent lead to profit as surely as haste leads to poverty By J. K. Kiingati Page 1
Time is precious, but we do not know yet how precious it really is. We will only know when we are no longer able to take advantage of it…
1. RANDOM VARIABLES
1.1 Introduction
In application of probability, we are often interested in a number associated with the outcome
of a random experiment. Such a quantity whose value is determined by the outcome of a
random experiment is called a random variable. It can also be defined as any quantity or
attribute whose value varies from one unit of the population to another.
A discrete random variable is function whose range is finite and/or countable, Ie it can only
assume values in a finite or countably infinite set of values. A continuous random variable is
one that can take any value in an interval of real numbers. (There are uncountably many real
numbers in an interval of positive length.)
1.2 Discrete Random Variables and Probability Mass Function

Consider the experiment of flipping a fair coin three times. The number of tails that appear is
noted as a discrete random variable. X= number of tails that appear in 3 flips of a fair coin.
There are 8 possible outcomes of the experiment: namely the sample space consists of
S  HHH , HHT , HTH , HTT , THH , THT , TTH , TTT
X  0 1, 1, 2 , 1, 2, 2, 3
are the corresponding values taken by the random variable X.
Now, what are the possible values that X takes on and what are the probabilities of X taking a
particular value?
From the above we see that the possible values of X are the 4 values
X = 0, 1, 2, 3
Ie the sample space is a disjoint union of the 4 events {X = j } for j=0,1,2,3
Specifically in our example:
X  0 = HHH X  1 = HHT, HTH, THH
X  2 = TTH, HTT, THT X  3 = TTT
Since for a fair coin we assume that each element of the sample space is equally likely (with
probability 18 , we find that the probabilities for the various values of X, called the probability
distribution of X or the probability mass function (pmf). can be summarized in the following
table listing the possible values beside the probability of that value
X 0 1 2 3
1 3 3 1
P(X=x) 8 8 8 8
Note: The probability that X takes on the value x, ie p(X  x) , is defined as the sum of the
probabilities of all points in S that are assigned the value x.
We can say that this pmf places mass 83 on the value X = 2 .
The “masses” (or probabilities) for a pmf should be between 0 and 1.
The total mass (i.e. total probability) must add up to 1.
Definition: The probability mass function of a discrete variable is a graph, table, or formula
that specifies the proportion (or probabilities) associated with each possible value the random
variable can take. The mass function P(X  x) (or just p(x) has the following properties:
0  p(x)  1 and  p(x)  1
all x
More generally, let X have the following properties
i) It is a discrete variable that can only assume values x1 , x2 , .... xn
ii) The probabilities associated with these values are
P( X  x1 )  p1 , P( X  x2 )  p2 ……. P( X  xn )  pn
n
Then X is a discrete random variable if 0  pi  1 and p
i 1
i 1
Remark: We denote random variables with capital letters while realized or particular values
are denoted by lower case letters.
Example 1
Two tetrahedral dice are rolled together once and the sum of the scores facing down was noted.
Find the pmf of the random variable ‘the sum of the scores facing down.’
Solution
+ 1 2 3 4 Therefore t is given the pmf by the table
1 2 3 4 5 below
2 3 4 5 6 x 2 3 4 5 6 7 8
1 1 3 1 3 1 1
3 4 5 6 7 P(X=x) 16 8 16 4 16 8 16
4 5 6 7 8 This can also be written as a function

 x161 for x  2 , 3, 4 , 5
X = 1, 2, 3, 4 , 5 , 6 , 7 , 8 P( X  x )   9 x
 16 for x  6 , 7, 8
Example 2
The pmf of a discrete random variable W is given by the table below
w -3 -2 -1 0 1
P(W=w) 0.1 0.25 0.3 0.15 D
Find the value of the constant d, P 3  w  0 , Pw  -1 and P 1  w  1
Solution
 p(W  w)  1  0.1  0.25  0.3  0.15  d  1  d  0.2
all w
P 3  w  0  P(W  3)  P(W  2)  P(W  1)  0.65

Pw  -1  Pw  0  Pw  1  0.15  0.2  0.35
P 1  w  1  P(W  0)  0.15
Example 3
A discrete random variable Y has a pmf given by the table below
Y 0 1 2 3 4
P(Y=y) c 2c 5c 10c 17c
Find the value of the constant c hence computes P1  Y  3
Solution
 p(Y  y)  1  c(1  2  5  10  17)  1  c  351
ally
P1  Y  3  P(Y  1)  P(Y  2)  352  355  15
Exercise 1.1
1. A die is loaded such that the probability of a face showing up is proportional to the face
number. Determine the probability of each sample point.
2. Roll a fair die and let X be the square of the score that show up. Write down the
probability distribution of X hence compute P X  15 and P3  X  30
3. Let X be the random variable the number of fours observed when two dice are rolled
together once. Show that X is a discrete random variable.
4. The pmf of a discrete random variable X is given by P( X  x )  kx for x  1, 2 , 3, 4 , 5 , 6
Find the value of the constant k, P X  4 and P3  X  6
5. A fair coin is flip until a head appears. Let N represent the number of tosses required to
realize a head. Find the pmf of N
6. A discrete random variable Y has a pmf given by P(Y  y )  c 34  for y  0 ,1, 2 , .....
x
Find the value of the constant c and PX  3

2x
7. Verify that f(x)  for x  0 ,1, 2 , .....k can serve as a pmf of a random variable X.
k (k  1)
8. For each of the following determine c so that the function can serve as a pmf of a random
variable X.
a) f(x)  cx for x 1, 2 , 3, 4 , 5 c) f(x)  c 16  for x  0 ,1, 2 , 3.....
x
b) f(x)  cx 2 for x  0 ,1, 2 , .....k d) f(x)  c2 x for x  for x  0 ,1, 2 , ....
9. A coin is loaded so that heads is three times as likely as the tails. For 3 independent
tosses of the coin find the pmf of the total number of heads realized and the probability of
realizing at most 2 heads.
1.3 Continuous Random Variables and Probability Density Function

A continuous random variable can assume any value in an interval on the real line or in a
collection of intervals. The sample space is uncountable. For instance, suppose an experiment
involves observing the arrival of cars at a certain period of time along a highway on a
particular day. Let T denote the time that lapses before the 1st arrival, the T is a continuous
random variable that assumes values in the interval [0 , )
Definition: A random variable X is continuous if there exists a nonnegative function f so that, for
every interval B, P X  B    f(x) dx. The function f = f(x) is called the probability density
B
function of X.
Definition: Let X be a continuous random variable that assumes values in the interval
( , ) , The f(x) is said to be a probability density function (pdf) of X if it satisfies the
following conditions
b 
f(x)  0 for all x , pa  x  b    f(x) dx  1 and  f(x) dx  1
a -
The support of a continuous random variable is the smallest interval containing all values of x
where f(x) >= 0.
Remark A crucial property is that, for any real number x, we have P(X  x)  0 (implying
there is no difference between P(X  x) and P(X  x) ); that is it is not possible to talk about
the probability of the random variable assuming a particular value. Instead, we talk about the
probability of the random variable assuming a value within a given interval. The probability
of the random variable assuming a value within some given interval from x  a to x  b is
defined to be the area under the graph of the probability density function
between x  a and x  b .
Example 1
Let X be a continuous random variable. Show that the function
 1 x, 0  x  2
f(x)   2 is a pdf of X hence compute P0  X  1 and P 1  X  1
 0, elsewhere
Solution
x 
2

2 2
f(x)  0 for all x in the interval 0  x  2 and 1
2
xdx  1
4 0  1 . Therefore f(x)is indeed a
0
pdf of X.
x 
1
Now P0  X  1   12 xdx  1
4
2 1
0  1
4
and
0
P 1  X  1  P 1  X  0  P0  X  1  0  14  14
Example 2
The time X, in hours, between computer failures is a continuous random variable with density
e 0.01x for x  0
f(x)   Find  hence compute P50  X  150 and PX  100
0 , elsewhere
Solution
f(x)  0 for all x in 0  x   Thus 1    e 0.01x dx  100 e 0.01x

0
 
0  100    0.01..
Now P50  X  150  0.01 e 0.01x dx   e 0.01x

150
50
 
150
50  e 0.5  e 1.5  0.3834005 and
PX  100  0.01 e 0.01x dx   e 0.01x

100
0
 
100
0  1  e 1  0.6321206
Example 3 A continuous random variable X has a probability density function given by

0.25 , 0  x  2

f(x)  0.5 x  c , 2  x  3 Find c hence compute P1  X  2.5 .
 0 , elsewhere

Solution
12 x  cdx  12  54  c
2 3
all x
f(x)dx  1   1
0 4
dx  
2
 c   34
P1  X  2.5  P1  X  2  P2  X  2.5   12 x  34 dx  14  x 4_ 3x 2

2 2.5 2.5
dx  
2
1
 14  163  167
1 4 2
Exercise 1.2
cx, 0  x  1
1) Suppose that the random variable X has p.d.f. given by f(x)   Find the
0, elsewhere
value of the constant c hence determine m so that PX  m  12
 5x  k , 0  x  3
2) Let X be a continuous random variable with pdf f(x)   Find the value
 0, elsewhere
of the constant k hence compute P1  X  3
k (1  y ), 4  x  7
3) A continuous random variable Y has the pdf given by f(y)   Find
 0, elsewhere
the value of the constant k hence compute PY  5 and P5  Y  6
4) The continuous random variable X has probability density function
k ( x  1)2 , 0  x  2
f( x)   Find the value of the constant k k hence compute P X  1.5
0, otherwise
k (1  y 2 ),  1  x  1
5) A continuous r.v Y has probability density function f( x)   Find
0, otherwise
the value of the constant k hence compute PY  0.5 and P(0.5  Y  0.5)
kx(1  x) , 0  x  1
6) Let X be a continuous r.v witha pdf f ( x)   . Calculate P X  12  14  
0, otherwise
6 , 0  x  2
 1
7) The pdf of a random variable X is given by f ( x)   1

3 , 2  x  4

Sketch the graph of f(x) hence find P(1  x  3)
k (1  x) 2 ,  2  x  0

8) A continuous random variable X has the pdf given by f( x)  4k 0  x  43 Find
0, elsewhere

the value of the constant k hence compute P X  1 and P 1  X  1
 kx, 0  x  2

9) A continuous random variable X has the pdf given by f(x)  k 4  x  2  x  4 Find

 0 , elsewhere
the value of the constant k hence compute P X  3 and P1  X  3
10) The length of time in minutes a customer queue in a post office is a random variable T with pdf
 
c 81  t 2 , 0  t  9
f(t )  
0, elsewhere
a) Find the value of the constant c and the cdf of T
b) Find the probability that a customer will wait for more than 3 minutes
c) A customer has been queuing for 3 minutes, find the probability that this customers will be
queuing for at least 7 minutes
d) Three customers are selected at random, find the probability that exactly 2 of them had to
queue for longer than 3 min
1.4 Distribution Function of a Random Variables

Definition: For any random variable X, we define the cumulative distribution function (CDF), F(x)
 x
  f(t) If X is discrete
as F(x)  P X  x   t   for every x.
 f(t)dt If X is continuous
 
x
Here t is introduced to facilitate summation /integration//
Properties of any cumulative distribution function

 lim F(x)  1 and lim F(x)  0
x  x 
 F(x) is a non-decreasing function.
 F(x) ) is a right continuous function of x. In other words lim F(t)  F(x)
tx
Reminder If the cdf of X is F(x) and the pdf is f(x) , then differentiate F(x) to get f(x), and
integrate f(x) to get F(x) ;
Theorem: For any random variable X and real values a < b, Pa  X  b  F(b) - F(a)
Example 1
 201 1  x  for x 1, 2 , 3 , 4 , 5
Let X be a discrete random variable with pmf given by f( x)   .
 0 , elsewhere
Determine the cdf of X hence compute P X  3
Solution
 4  x  1  x( x403)
x x
F(x)   f(t) 
t  
1
20  (t  1) 
t 1
1
20
(2  3  ....  x)  1 x
20 2
0 for x 1
 x  3)
F( x)   x ( 40 for x 1, 2 , 3 , 4 , 5 Recall for an AP S n n
2 2a  n  1d 
1 for x  5

P X  3  1  P X  3  1  340
6
 11
20
Example 2
Suppose X is a continuous random variable whose pdf f(x) is given by
 1 x, 0  x  2
f( x)   2 . Obtain the cdf of X hence compute P X  23 
 0, elsewhere
Solution
0, x  0
   2
x x
F(x)   f(t)dt   12 tdt  14 t 2 0  14 x 2 thus F( x)   x4 , 0  x  2
x
- 0 1, x  2

P X  3   1  P X  3   1  4  3   9
2 2 1 2 2 8
Example 3
A continuous random variable X has a probability density function given by
0.25 , 0  x  2

f( x)  0.5 x  c , 2  x  3 Find the cdf of X hence compute P1.5  X  2.5 .
 0 , elsewhere

Solution
 x 1 dt  x
0 4
x
4
F(x)   f(t)dt   x
  12 t - 34 dt  k  x 4-3x  12  k
2
-
2
Under the two levels, F(2) must be the same. (reason for introducing k)
0 for x  0
x
 4 for 0  x  2
 F(2)  2  k  F( x)   2
1
 4  1 for 2  x  3
x -3x
1, for x  2

Exercise 1.3
x3  k
1. The CDF of a discrete random variable X is given by F( X )  , x  1, 2, 3
40
a) Show that k=13 b) Find the pmf of x

C , 0  x  4
2. The pdf of a continuous random variable X is given by f( x)   x Find the
 0, elsewhere
value of the constant C, the cdf of X and P X  1
kx(1  x), 0  x  1
3. The pdf of a random variable X is given by g( x)   Find the value
 0, elsewhere
of the constant k, the cdf of X and the value of m such that G(x)  12
4. A continuous random variable x has b) Find the cdf of x hence compute
cumulative distribution function P(0.5  x  1.5)
0), x  2 6. A continuous random variable X has

F( x)   16 x  13 ,  2  x  4 Find the pdf f(x) given by
 1, x  4  x2

 , 0 x3
probability density function f(x) of X.  45
5. A random variable X has pdf 0.2 , 3  x  4
f( x)  
2 , 0  x 1
1
10  x , 4  x  10
 1  30
f( x)   x  2 1  x  k where k is a
0 , otherwise 0 , elsewhere
 
constant. Sketch f(x). Also find the cdf of x and
a) Sketch a graph of f(x) hence find k. hence compute p( X  8)
9
7. If the cdf of a random variable Y is given by F( y )  1  for Y  3 and
y2
F(x)  0 for Y  3 , find P X  5 , P X  8 and the pdf of X
8. A random variable X has a cdf F(X)  2 x2  x 4 for x  0,1 . Compute P( 14  x  34 ) and
also find the pdf of x
9. Find the cdf of a random variable Y whose pdf is given by;
x 2 , 0  x  1
 13 , 0  x  1 1 , 1  x  2
  2
a) f( x)   13 , 2  x  4 b) f( x)  (3 x )
 0, elsewhere  2, 2 x3
  0, elsewhere
10. The continuous random variable x has probability density function f(x) given by
k ( x 2  2 x  2), 0  x  3

f(X)  3k 3 x  4 where k is a constant. Show that k  19
 0, elsewhere

a) Find the cumulative distribution function F(x) and the mean of X.
b) Show that the median of X lies between x = 2.6 and x = 2.7
11. The lifetime, X, in tens of hours, of a battery has a cumulative distribution function F(x)
0, x  1

given by F( x)   94 ( x 2  2 x  3) , 1  x  1.5
 1, x  1.5

a) Find the median of X, giving your answer to 3 significant figures.
b) Find, in full, the probability density function of the random variable X.

c) Find P(X> 1.2)
12. The continuous random variable T represents the time in hours that students spend on
homework. The cumulative distribution function of T is
0, t  0

F(t )  k (2t 3  t 4 ) , 0  t  1.5 where k is a positive constant.
 1, t  1.5

a) Show that k  16 27 hence and find the probability density function f(t) of T.
b) Find the proportion of students who spend more than 1 hour on homework.
c) Show that E(T ) = 0.9 and F(E(T )) = 0.4752.
d) A student is selected at random. Given that the student spent more than the mean
amount of time on homework, find the probability that this student spent more than 1
hour on homework.
1.5 Derived Random Variables

Give the pdf of a random variable say X, we can obtain the distribution of a second random
variable say Y provided that we know some functional relationship between X and Y say
Y  u( X ). Eg Y  2 X  3 , Y  X 3 , Y  X  c etc
For a 1-1 relationship between X and Y eg Y  2 X  3, f(x )and g(y) yields exactly the same
probabilities only the random variable and the set of values it can assume changes.
Example 1
 x 6 for x 1, 2 , 3
Give the pmf of a random variable X as f( x)   find the pmf of Y  X 2
0 , elsewhere
Solution
The only values of Y with non-zero probabilities are Y = 1, Y = 4 and Y = 9 . Now
PY = 1  PX2 = 1  PX = 1  16 PY = 4  P X2 = 4  PX = 2  13 and 
 
PY = 9  P X2 = 9  PX = 3  12
In some cases several values of X will give rise to the same value of Y. The procedure
is just the same as above but it is necessary to add the several probabilities that are associated
with each value x that provides a unique value y.
 x 1
 for x  0 , 1, 2 , 3 , 4
Example 2 Give the pmf of a r.v X as f( x)   15 find the pmf of
 0 , elsewhere
Y   X  2
2
Solution
x 0 1 2 3 4
y 4 1 0 1 4
PY = 0  PX = 2  15 PY = 1  PX = 1  PX = 3  152  154  52

PY = 1  PX = 0  PX = 4  151  13  52 Therefore the pmf of Y can be written as
Y 0 1 4
PY = y 2
5
2
5
2
5
\Exercise 1.4
 1 6 for x 1, 2 , 3 , 4 , 5 , 6
1. Suppose the pmf of a r.v X is given by f( x)   , Obtain the pmf of
0 , elsewhere
Y  2 X 2 and Z  X  3
 x2  1
 for x  0 , 1, 2 , 3
2. Let the pmf of a r.v X be given by f( x)   18 , determine the pmf of
 0 , elsewhere

Y  X 1
2
 x 10 for x  0 , 1, 2 , 3 , 4
3. Suppose the pmf of a r.v X is given by f( x)   , Obtain the pmf of
 0 , elsewhere
Y X 2
 1  for x  1, 2 , 3 , ....
 x
4. Let the pmf of a r.v X be given by f( x)   2 , determine the pmf of


 0 , elsewhere
1 if x is odd
Y
0 if x is even
 16  56  for x  0 ,1, 2 , 3 , ....
 x
5. Suppose the r.v X has a pmf given by f( x)   , Obtain the pmf


 0 , elsewhere
of Y  X  1
6. Let X be a discrete r.v with pmf as tabulated below. Find the pmf of Y  ( X  2) 2
X -2 -1 0 1 2
1 1 1 1 1
P(X=x) 4 8 8 4 4
1.6 Change of Variable Technique

dx
.Let X be c. r.v with pdf f(x) and let Y be a function of X, then the pdf of Y g(y)  f(x)
dy
NB If F(x) is the cdf of a r.v X, then a r.v Y  F(x) has a uniform distribution over 0 , 1
Example 1
5 x 4 , 0  x  1
A continuous r.v X has a pdf given by f( x)   . Determine the pdf of Y  x 3
 0 otherwise
Solution
1 2 
  5 y 3 , 0  x  1
2
dx 1  2 3 dx 4
Yx  X Y   Y g(y)  f(x)  5 y 3  y 3  3
3 1 1
3
dy 3 dy 3 
 0 otherwise
24 x 2 , 0  x  12
Example 2 A r.v X has pdf f( x)   determine the pdf of Y  8 X 3
 0 otherwise
Solution
Y  8 x 3  X  12 Y 3 
1 dx 1  2 3
6y g(y)  f(x)
dx 2
 24 12 y 3  16 y  3  
1 2

1, 0  x  1

dy dy 0, elsewhere
NB Y  8 X 3 is the cdf of X
Exercise 1.5
5 x 4 , 0  x  1
1. For a r.v X with f( x)   , determine the pdf of Y  2 ln x and its range.
 0 otherwise
e  x , x  0
2. A r.v X has pdf f( x)   determine the pdf of Y  X 4
0 otherwise
3. Let X be a continuous random variable with density function f( x)  x12 for x  2 and
f( x)  0 otherwise . Determine the density function of Y  1
X 1
.
 2
  X 
4. The probability density function of X is given by f( x)   ( x  4)
2
Obtain
 0 otherwise

the probability density function of Y  tan1 ( 2x )
5. Which transformation will change a r.v X with pdf is as below to a uniform R.V Y whose
2e 2 x , x  0  1 (x - 3), 3  x  5
range is 0  x  1 a) f( x)   b) f( x)   2
 0 otherwise  0 otherwise
 
 for x  1
6. Suppose that X has probability density function f( x)   x ln  If U  X 2 , what
 0 , elsewhere
is the probability density function f(u) of U?
1.7 Expectation and Variance of a Random Variable
1.7.1 Expected Values

One of the most important things we'd like to know about a random variable is: what value
does it take on average? What is the average price of a computer? What is the average value
of a number that rolls on a die? The value is found as the average of all possible values,
weighted by how often they occur (i.e. probability)
Definition: Let X be a random variable with probability distribution p( X  x) . Then the
expected value of X, denoted E (X ) or  , is given by;
 
  xp ( X  x) if X is discrete
E( x)     x   .
  xp ( X  x)dx if X is continuous

Theorem: Let X be a r.v. with probability distribution p(X=x) and let g(x) be a real-valued
function of X. ie , then the expected value of g(x) is given by
 
  g ( x) p( X  x) if X is discrete
Eg( x)   x   .
 g ( x) p( X  x)dx if X is continuous


Theorem: Let X be a r.v. with probability distribution p( X  x) . Then
(i) E(c) = c, where c is an arbitrary constant.

(ii) Eax  b  a  b where a and b are arbitrary constants
(iii) Ekg(x)   kEg(x)  where g(x) is a function of X
n  n
(iv) Eag1 (x)  bg 2 (x)   aEg1 (x)   bEg 2 (x)  and in general E  ci g i (x)    ci Eg i (x) 
 i 1  i 1
are functions of X. This property of expectation is called linearity property
Proof
We will sketch the proof using a continuous random variable since the proof using the
discrete random variable is similar and was also discussed in probability and statistics I.
(i) Ec   cP ( X  x)dx  c  P( X  x)dx  c(1)  c
All x All x
(ii) Eax  b   (ax  b) P( X  x)dx   axP( X  x)dx   bP( X  x)dx

All x All x All x
a  xP( X  x)dx  b  P( X  x)dx  a  b

All x All x
(iii) Ekg(x)    kg( x) P( X  x)dx  k  g ( x) P( X  x)dx  kEg(x) 

All x All x
(iv) Eag1 (x)  bg 2 (x)    ag (x)  bg (x) P( X  x)dx   ag (x) P( X  x)dx   bg (x) P( X  x)dx
1 2 1 2
All x All x All x
 Eag1 (x)   Ebg 2 (x)   aEg1 (x)   bEg 2 (x)  ie from part iii
1.7.2 Variance and Standard Deviation

Definition: Let X be a r.v with mean E (X )   , the variance of X, denoted  2 or Var(X) , is
given by Var ( X )   2  E( X   ) 2 . The units for variance are square units. The quantity that has
the correct units is standard deviation, denoted  . It’s actually the positive square root
of Var(X) .
  Var ( X )  E ( X   ) 2 .
Theorem: Var ( X )  E ( X   ) 2  E ( X ) 2   2
Proof:
 
Var ( X )  E( X   )2  E X 2  2 X   2  E( X )2  2E( X )   2  E( X )2   2
Theorem: Var (aX  b)  a var( X ) 2
Proof:
Recall that EaX  b  a  b therefore

Var (aX  b)  E(aX  b)  (a  b)  Ea X     E a 2  X     a2 E  X     a 2 var( X )
2 2 2
  2

Remark
(i) The expected value of X always lies between the smallest and largest values of X.
(ii) In computations, bear in mind that variance cannot be negative!
Example 1
Given a probability distribution of X as below, find the mean and standard deviation of X.
x 0 1 2 3
P(X=x) 1/8 1/4 3/8 1/4
Solution
3
X 0 1 2 3 total E ( X )     xp ( X  x)  1.75 and
p ( X  x) 1
8
1
4
3
8
1
4 1 x 0
xp ( X  x) 0 1
4
3
4
3
4
7
4
standard deviation
x p ( X  x)
2
0 1
4
3
2
9
4 4   E ( X 2 )   2  4  1.752  0.968246
Example 2 The probability distribution of a r.v X is as shown below, find the mean and
standard deviation of; a) X b) Y  12 X  6 .
x 0 1 2
P(X=x) 1/6 1/2 1/3
Solution
2
X 0 1 2 total E ( X )     xp ( X  x)  7 6 and
p ( X  x) 1
6
1
2
1
3 1 x 0
xp ( X  x) 2
E ( X 2 )   x 2 p( X  x)  116
1 2 7
0 2 3 6
x p ( X  x)
2
0 1
2
4
3
11
6 x 0
Standard deviation   E ( X 2 )   2  11
6  (7 6) 2  17
6  1.6833
Now E(Y )  12E( X )  6  12(7 6 )  6  20
Var (Y )  Var (12 X  6)  122  Var ( X )  144  17
6  242.38812
 1 x, 0  x  2
Example 3 A continuous r.v X has a pdf given by f(x)   2 , find the mean
 0, elsewhere
and standard of X
Solution
   
2 2
 4 
E(x)   xf( x)dx   2 x dx  6 x 0  and E(x )   x f( x)dx   12 x 3dx  18 x 4 0  2
2 2
1 2 1 3 2 2
 3  
0 0
Standard deviation   E ( X 2 )   2  2  ( 4 3 )2  3
2
Exercise 1.6
1. Suppose X has a probability mass function given by the table below
x 2 3 4 5 6
P(X=x) 0.01 0.25 0.4 0.3 0.04
Find the mean and variance of; X
2. Suppose X has a probability mass function given by the table below
x 11 12 13 14 15
P(X=x) 0.4 0.2 0.2 0.1 0.1
Find the mean and variance of; X
3. Let X be a random variable with P(X = 1) = 0.2, P(X = 2) = 0.3, and P(X = 3) = 0.5. What is the
expected value and standard deviation of; a) X b) Y  5 X  10 ?
4. A random variable W has the probability distribution shown below,
w 0 1 2 3
P(W=w) 2d 0.3 d 0.1
Find the values of the constant d hence determine the mean and variance of W. Also find
the mean and variance of Y  10 X  25
5. A random variable X has the probability distribution shown below,

x 1 2 3 4 5
P(X=x) 7c 5c 4c 3c c
Find the values of the constant c hence determine the mean and variance of X.
6. The random variable Z has the probability distribution shown below,
z 2 3 5 7 11
1 1 1
P(Z=z) 6 3 4 x y
If E ( Z )  4 23 , find the values of x and y hence determine the variance of Z
 m , m  1, 2 , 3 , ...,8
7. A discrete random variable M has the probability distribution f(m)   36 ,
 0 , elsewhere
find the mean and variance of M
 510
y
, y  1, 2 , 3 , 4
8. For a discrete random variable Y the probability distribution is f(y)   ,
 0 , elsewhere
calculate E (Y ) and var(Y)
kx for x  1, 2 , 3 , 4
9. Suppose X has a pmf given by f(x)   , find the value of the constant k
 0 , elsewhere
hence obtain the mean and variance of X
kx , x  2, 4, 6

10. A discrete random variable X has a probability function P( X  x)  k x  2, x  8
0, elsewhere

a) Show that k  181 hence find E (X ) and E ( X 2 )
b) Calculate var3  4 X  , give your answer to 3 significant figures.
11 A biased die with six faces is rolled. The discrete random variable X represents the score
of the uppermost face. The probability distribution of X is shown in the table below
X 1 2 3 4 5 6
P(X=x) a a a b b 0.3
a) Given that E(X) = 4.2, find the values of a and b.
 
b) Show that E X 2  20.4 c) Find var 5  3x 
k 1  x  , x  1,0,1,2 2
12 A discrete random variable X has probability function P( X  x)  
0, elsewhere
a) Show that k  16 c) Show that E ( X 2 )  43
b) Find E (X ) d) Find var 1  3x 
13 A team of 3 is to be chosen from 4 girl and 6 boys. If X is the number of girls in the
team, find the probability distribution of X hence determine the mean and variance of X
14 A fair six sided die has; ‘1’ on one face, ‘2’ on two of its faces and ‘3’ on the remaining
three faces. The die is rolled twice. If T is the total score write down the probability
distribution of T hence determine; a) the probability that T is more than 4. b) the mean
and variance of T
kr for 0  r  4
15 The pdf of a continuous r.v R is given by f(r)   , (a) Determine c. hence
 0 , elsewhere
Compute P(1 0  r  2) , E(X) and Var(X).
k (1  10
m
) for M  10
16 A continuous r.v M has the pdf given by f(m)   , find the value of
 0 , elsewhere
the constant k, the mean and the variance of X
k (1  x) for 0  x  1
17 A continuous r.v X has the pdf given by f(x)   , findt the value of
 0 , elsewhere
the constant k. Also find the mean and the variance of X
 d 2 if T  1
18 The lifetime of new bus engines, T years, has continuous pdf f(t)   t find the
 0 , if T  1
value of the constant d hence determine the mean and standard deviation of T
19 An archer shoots an arrow at a target. The distance of the arrow from the centre of the
k (3 + 2x - x 2 ) if x  3
target is a random variable X whose p.d.f. is given by f(x)   find
 0 , if x  3
the value of the constant k. Also find the mean and standard deviation of X
20 The random variable Y has probability density function f(y) given by
ky(a  y ), 0  x  3
f(y)   where k and a are positive constants.
 0, elsewhere
2
a) Explain why a ≥ 3 and then show that k 
9(a  2)
b) Given that E(Y) = 1.75 , show that a = 4 and write down the value of k.
c) For these values of a and k, sketch the probability density function,
d) Write down the mode of Y.
a  bx , 0  x  5
21 A continuous random variable x has the following pdf f( x)   where a and b
0 , otherwise
are constants. Show that 10a  25b  2
a) Given E  X  
35
, find a second equation in a and b hence find the values of a and b.
12
b) Find the median of X
22 The queuing time X minutes of a customer at a till of a supermarket has a pdf
 323 x(k  x) , 0  x  k
f( x)  
0 , otherwise
a) Show that k  4 . Also find E (x) and var(x)
b) Find the probability that a randomly chosen customers queuing time will differ from the mean
by at least half a minute
23 The probability density function f(x) can be written in the following form.
ax , 0  x  2

f( x)  b  ax , 2  x  4
0 , otherwise

a) Find the values of the constants a and b.
b) Show that σ, the standard deviation of X, is 0.816 to 3 decimal places.
c) Find the lower quartile of X.
d) State, giving a reason, whether P(2 – σ < X < 2 + σ) is more or less than 0.5
k (1  x) ,  1  x  0

24 A continuous r.v X has the pdf given by f(x)  2k (1  x), 0  x  1 , find the value of
 0 , elsewhere

the constant k. Also find the mean and the variance of X
e  x for x  0
25 A continuous r.v X has the pdf given by f(x)   , find the mean and
 0 , elsewhere
3x
standard deviation of; a) X b) Y  e 4
1.8 Mode, Median, Quartiles and Percentiles

Another measure commonly used to summarize random variables are the mode and median;
Mode is the value of x that maximizes the pdf. That is the value of x for which f | (x) = 0 .
Median is the value m such that “half of the distribution lies to the left of m and half to the
right”. More formally, m should satisfy FX (m) = 0.5 .
Note: If there is a value m such that the graph of y= f(x) is symmetric about x=m, then both
the expected value and the median of X are equal to m.
The lower quartile Q l and the upper quartile Q 3 are similarly defined by
FX (Ql ) = 0.25 and FX (Q3 ) = 0.75
Thus, the probability that X lies between Q l and Q 3 is 0.75 - 0.25 = 0.5 , so the quartiles give
an estimate of how spread-out the distribution is. More generally, we define the nth percentile
of X to be the value of xn such that FX (x n ) = 0.01n or n 100 , that is, the probability that X is
smaller than xn is n%.
2 x, 0  x  1
Example A random variable X has the pdf given by f(x)   Find the lower,
 0 , elsewhere
middle and upper quartiles.
Solution
On the interval 0  x  1, the cdf of X is given by F(x)  x 2 thus
a) At lower quartile Q l , F(Ql )  Ql  0.25  Ql  0.25  0.5
2
b) At median m, F(m)  m2  0.5  m  0.5  1

2
c) At upper quartile u F(Q3 )  Q3  0.75  Q3  0.75 

2 3
2
Qn Find the 64th percentile of the pdf in the above example.
k (1  x) for 0  x  1
A continuous r.v X has the pdf given by f(x)   , find the mode.
 0 , elsewhere
 12 e  2x for x  0
Let X be a continuous random variable with density function f(x)   .
 0 , elsewhere
Determine the 25th percentile of the distribution of X.
2. PROBABILITY DISTRIBUTION
2.1 Discrete Distribution
Among the discrete distributions that we will discuss in this topic includes the Bernoulli,
binomial, Poisson, geometric and hyper-geometric
2.1.1 Bernoulli distribution

Definition: A Bernoulli trial is a random experiment in which there are only two possible
outcomes - success and failure. Eg
 Tossing a coin and considering heads as success and tails as failure.
 Checking items from a production line: success = not defective, failure = defective.
 Phoning a call centre: success = operator free; failure = no operator free.
A Bernoulli random variable X takes the values 0 and 1 and P( X  1)  p and
P( X  0 )  1  p
Definition: A r.v X is said to be a real Bernoulli distribution if it’s pmf is given by;
 p x (1  p)1 x for x  0 ,1
P( X  x )  
 0 otherwise
We abbreviate this as X ~ B(p) ie p is the only parameter here. It can be easily checked that
the mean and variance of a Bernoulli random variable are   p and  2  p(1  p1)
2.1.2 Binomial Distribution

Consider a sequence of n independent, Bernoulli trials, with each trial having two possible
outcomes, success or failure. Let p be the probability of a success for any single trial. Let X
denote the number of successes on n trials. The random variable X is said to have a binomial
distribution and has probability mass function
P( X  x )n Cr  p x (1  p)n  x for x  0 ,1, 2 ..... n
We abbreviate this as X ~ Bin ( n , p) read as “X follows a binomial distribution with
parameters n and p ”. n Cr Counts the number of outcomes that include exactly x successes
and n  x failures.
The mean and variance of a Binomial random variable are respectively given by;
  np and  2  np(1  p1)
n
Let’s check to make sure that if X has a binomial distribution, then  P( X  x )  1 . We will
x 0
need the binomial expansion for any polynomial:
n n
 p  q n   n Cr  p x q n  x therefore  n Cx  p x (1  p)n  x   p  1  p n  1n  1
x 0 x 0
Example 1
A biased coin is tossed 6 times. The probability of heads on any toss is 0:3. Let X denote the
number of heads that come up. Calculate: (i) P( X  2 ) (ii) P( X  3) (iii) P(1  X  5 )
Solution
If we call heads a success then X has a binomial distribution with parameters n=6 and p=0:3.
(i) P( X  2 )6 C2  (0.3)2 (0.7)4  0.324135
(ii) P( X  3 )6 C3  (0.3)3 (0.7)3  0.18522
(iii) P(1  X  5 )  P( X  2 )  P( X  3 )  P( X  4 )  P( X  5 )
 0.324  0.185  0.059  0.01  0.578
Example 2 A quality control engineer is in charge of testing whether or not 90% of the
DVD players produced by his company conform to specifications. To do this, the engineer
randomly selects a batch of 12 DVD players from each day's production. The day's
production is acceptable provided no more than 1 DVD player fails to meet specifications’.
Otherwise, the entire day's production has to be tested.
a) What is the probability that the engineer incorrectly passes a day's production as
acceptable if only 80% of the day's DVD players actually conform to specification?
b) What is the probability that the engineer unnecessarily requires the entire day's
production to be tested if in fact 90% of the DVD players conform to specifications?
Solution
Let X denote the number of DVD players in the sample that fail to meet specifications.
a) In part a we want P( X  1) with binomial parameters n  12 and p  0.2
P( X  1)  P( X  0 )  P( X  1) 12 C0  (0.2) 0 (0.8)12 12 C1  (0.2)1 (0.8)11
 0.069  0.206  0.275
b) In part b we require P( X  1)  1  P( X  1) with parameters n  12 and p  0.1 .
P( X  1)  P( X  0 )  P( X  1) 12 C0  (0.1)0 (0.9)12 12C1  (0.1)1 (0.9)11  0.659
So P(X  1)  0.341
Example 3 Bits are sent over a communications channel in packets of 12. If the probability
of a bit being corrupted over this channel is 0.1 and such errors are independent, what is the
probability that no more than 2 bits in a packet are corrupted?
If 6 packets are sent over the channel, what is the probability that at least one packet will
contain 3 or more corrupted bits?
Let X denote the number of packets containing 3 or more corrupted bits. What is the
probability that X will exceed its mean by more than 2 standard deviations?
Solution
Let C denote the number of corrupted bits in a packet. Then in the first question, we want
P(C  2 )  P(C  0 )  P(C  1)  P(C  2 )
12 C0 (0.1)0 (0.9)12 12C1 (0.1)1 (0.9)1112C2 (0.1) 2 (0.9)10
 0.282  0.377  0.23  0.889.
Implying the probability of a packet containing 3 or more corrupted bits is
P(C  3 )  1  P(C  2 )  1 - 0.889 = 0.111.
Therefore X=’number of packets containing 3 or more corrupted bits” can be modelled with a
binomial distribution with parameters n  6 and p  0.111 . The probability that at least one
packet will contain 3 or more corrupted bits is:
P( X  1)  1  P( X  0 )  1 -6 C0  (0.111)0 (0.889)6  0.494.
The mean of X is E(X) = 6(0.111) = 0.666 and its standard deviation is
= 6(0.111) (0.889) = 0.77
So the probability that X exceeds its mean by more than 2 standard deviations is
P( X    2 )  P( X  2.2 )  P( X  3 ) since X is discrete.
Now P( X  3 )  1  P( X  2 )  1  P( X  0 )  P( X  1)  P( X  2 )

 1 - 6 C0  (0.111)0 (0.889)6  6 C1  (0.111)1 (0.889)5  6 C2  (0.111)2 (0.889)4 
 1 - (0.4936  0.3698  0.1026)  0.032
Exercise 2.1
1. A fair coin is tossed 10 times. What is the probability that exactly 6 heads will occur.
2. If 3% of the electric bulbs manufactured by a company are defective find the probability
that in a sample of 100 bulbs exactly 5 bulbs are defective.
3. Suppose that 10% of inmates in a large prison are known to be innocent. A non-profit
group randomly selects 20 inmates from this prison. Find the probability the group will
find at least 3 innocent inmates.
4. An oil exploration firm is formed with enough capital to finance 10 explorations. The
probability of a particular exploration being successful is 0.1. Find the mean and variance
of the number of successful explorations.
5. Emily hits 60% of her free throws in basketball games. She had 25 free throws in last
week’s game.
a) What is the expected number and the standard deviation of Emily’s hit?
b) Suppose Emily had 7 free throws in yesterday’s game, what is the probability that she
made at least 5 hits?
6. A coin is loaded so that heads has 60% chance of showing up. This coin is tossed 3 times.
a) What are the mean and the standard deviation of the number of heads that turned out?
b) What is the probability that the head turns out at least twice?
c) What is the probability that an odd number of heads turn out in 3 flips?
7. According to the 2009 current Population Survey conducted by the U.S. Census Bureau,
40% of the U.S. population 25 years old and above have completed a bachelor’s degree or
more. Given a random sample of 50 people 25 years old or above, what is expected
number of people and the standard deviation of the number of people who have
completed a bachelor’s degree.
8. Joe throws a fair die six times and face number 3 appeared twice. It he incredibly lucky or
unusual?
9. If the probability of being a smoker among a group of cases with lung cancer is .6, what’s
the probability that in a group of 8 cases you have; a) less than 2 smokers? b) More than
5? c) What are the expected value and variance of the number of smokers?
10. The manufacturer of the disk drives in one of the well-known brands of microcomputers
expects 2% of the disk drives to malfunction during the microcomputer’s warranty
period. Calculate the probability that in a sample of 100 disk drives, that not more than
three will malfunction
11. Manufacturer of television set knows that on an average 5% of their product is defective.
They sells television sets in consignment of 100 and guarantees that not more than 2 set
will be defective. What is the probability that the TV set will fail to meet the guaranteed
quality?
12. Suppose 90% of the cars on Thika super highways does over 17 km per litre.
a) What is the expected number and the standard deviation of cars on Thika super
highways that will do over 17 km per litre.in a random sample of 15 cars ?
b) What is the probability that in a random sample of 15 cars exactly 10 of these will do
over 17 km per litre?
2.1.3 Poisson distribution

Named after the French mathematician Simeon Poisson, the distribution is used to model the
number of events, (such as the number of telephone calls at a business, number of customers in
waiting lines, number of defects in a given surface area, airplane arrivals, or the number of
accidents at an intersection), occurring within a given time interval. Other such random events
where Poisson distribution can apply includes;
 the number of hits to your web site in a day
 the number of calls that arrive in each day on your mobile phone
 the rate of job submissions in a busy computer centre per minute.
 the number of messages arriving to a computer server in any one hour.
Poisson probabilities are useful when there are a large number of independent trials with a
small probability of success on a single trial and the variables occur over a period of time. It
can also be used when a density of items is distributed over a given area or volume. The
xe 
formula for the Poisson probability mass function is P( X  x)  , x  0 ,1, 2 , .... This is
x!
abbreviated as X ~ Po (  ) .  is the shape parameter which indicates the average number of
events in the given time interval. The mean and variance of this distribution are equal ie
 2  

 P( X  x )  1
Let’s check to make sure that if X has a poisson distribution, then x 0 . We will
 2
3

4
need to recall that e  1    
 ..... . Consequently
1! 4!2! 3!
 
x e   
x

x 0
P( X  x )  
x 0 x!
 e   e  e   e 0  1

x 0 x!
Remark The major difference between Poisson and Binomial distributions is that the Poisson
does not have a fixed number of trials. Instead, it uses the fixed interval of time or space in which
the number of successes is recorded.
Example 1 Consider a computer system with Poisson job-arrival stream at an average of 2 per
minute. Determine the probability that in any one-minute interval there will be
a) 0 jobs; b) exactly 3 jobs; c) at most 3 arrivals. d) more than 3 arrivals
Solution
Job Arrivals with =2
a) No job arrivals: P(X  0 )  e2  0.1353353
23 e 2
b) Exactly 3 job arrivals: P( X  3)   0.1804470
3!
c) At most 3 arrivals
 2 2 2 23 
d) P( X  3)  P( X  0)  P( X  1)  P( X  2)  P( X  3)  1    e  2  0.8571
 1 2 3! 
e) more than 3 arrivals P( X  3)  1  P( X  3) 1 - 0.8571  0.1429
Example 2 If there are 500 customers per eight-hour day in a check-out lane, what is the
probability that there will be exactly 3 in line during any five-minute period?
Solution
The expected value during any one five minute period would be 500 / 96 = 5.2083333. The
96 is because there are 96 five-minute periods in eight hours. So, you expect about 5.2
customers in 5 minutes and want to know the probability of getting exactly 3.
(-500/96) 3e-500/96
P( X  3)   0.1288 (approx)
3!
Example 3 If new cases of West Nile in New England are occurring at a rate of about 2 per
month, then what’s the probability that exactly 4 cases will occur in the next 3 months?
Solution
X ~ Poisson (=2/month)
(2 * 3) 4 e  ( 2*3) 6 4 e  ( 6)
P(X  4 in 3 months)    13.4%
4! 4!
Exactly 6 cases?
(2 * 3) 6 e  ( 2*3) 66 e  ( 6)
P(X  6 in 3 months)    16%
6! 6!
Exercise 2.2
1. Calculate the Poisson distribution whose λ (Average Rate of Success)) is 3 & X (Poisson
Random Variable) is 6.
2. Customers arrive at a checkout counter according to a Poisson distribution at an average
of 7 per hour. During a given hour, what are the probabilities that
a) No more than 3 customers arrive?
b) At least 2 customers arrive?
c) Exactly 5 customers arrive?
3. It is known from the past experience that in a certain plant there are on the average of 4
industrial accidents per month. Find the probability that in a given year will be less that 3
accidents.
4. Suppose that the change of an individual coal miner being killed in a mining accident
during a year is 1.1499. Use the Poisson distribution to calculate the probability that in
the mine employing 350 miners- there will be at least one accident in a year.
5. The number of road construction projects that take place at any one time in a certain city
follows a Poisson distribution with a mean of 3. Find the probability that exactly five road
construction projects are currently taking place in this city. (0.100819)
6. The number of road construction projects that take place at any one time in a certain city
follows a Poisson distribution with a mean of 7. Find the probability that more than four
road construction projects are currently taking place in the city. (0.827008)
7. The number of traffic accidents that occur on a particular stretch of road during a month
follows a Poisson distribution with a mean of 7.6. Find the probability that less than three
accidents will occur next month on this stretch of road. (0.018757)
follows a Poisson distribution with a mean of 7. Find the probability of observing exactly
three accidents on this stretch of road next month. (0.052129)
follows a Poisson distribution with a mean of 6.8. Find the probability that the next two
months will both result in four accidents each occurring on this stretch of road. (0.00985)
10. Suppose the number of babies born during an 8-hour shift at a hospital's maternity wing
follows a Poisson distribution with a mean of 6 an hour. Find the probability that five
babies are born during a particular 1-hour period in this maternity wing. (0.160623)
11. The average number of claims per day made to the Insurance Company for damage or
losses is 3.1. What is the probability that in any given day; (i)exactly 2 (ii) at most 2
(iii) more than 2 claims will be made?
12. The university policy department must write, on average, five tickets per day to keep
department revenues at budgeted levels. Suppose the number of tickets written per day
follows a Poisson distribution with a mean of 8.8 tickets per day. Find the probability that
less than six tickets are written on a randomly selected day from this distribution.
(0.128387)
13. A taxi firm has two cars which it hires out day by day. The number of demands for a car
on each day is distributed as Poisson distribution with mean 1.5. Calculate the proportion
of days on which neither car is used and the proportion of days on which some demands
is refused
14. If calls to your cell phone are a Poisson process with a constant rate =0.5 calls per hour,
what’s the probability that, if you forget to turn your phone off in a 3 hour lecture, your
phone rings during that time? How many phone calls do you expect to get during this
lecture?
15. The average number of defects per wafer (defect density) is 3. The redundancy built into
the design allows for up to 4 defects per wafer. What is the probability that the
redundancy will not be sufficient if the defects follow a Poisson distribution?
16. The mean number of errors due to a particular bug occurring in a minute is 0.0001
a) What is the probability that no error will occur in 20 minutes?
b) How long would the program need to run to ensure that there will be a 99.95% chance
that an error wills show up to highlight this bug?
Properties of Poisson
 The mean and variance are both equal to .
 The sum of independent Poisson variables is a further Poisson variable with mean
equal to the sum of the individual means.
 As well as cropping up in the situations already mentioned, the Poisson distribution
provides an approximation for the Binomial distribution.
2.1.4 Geometric Distribution

Suppose a Bernoulli trial with success probability p is performed repeatedly until the first
success appears we want to find the probability that the first success occurs on the yth trial. ie
let Y denote the number of trials needed to obtain the first success. The sample space
S={s;fs;ffs, fffs ffffs …}. This is an infinite sample space (though it is still discrete). What is
the probability of a sample point, say P( fffs)  P(Y  4) )? Since successive trials are
independent (this is implicit in the statement of the problem), we have
P( fffs)  P(Y  4)  q3p where q  1  p and 0  p  1
Definition: A r.v. Y is said to have a geometric probability distribution if and only if
 pq y 1for y  1, 2 , 3..... where q  1  p
P(Y  y)   .
0 otherwise
This is abbreviated as Y ~ Geo(p).
The only parameter for this geometric distribution is p (ie the probability of success in each
trial). To be sure everything is consistent; we should check that the probabilities of all the
sample points add up to 1. Now
 
p
y 1
P(Y  y )   pq y 1 
y 1 1 q
1
a
Recall sum to infinity of a convergent G.P is s 
1 r
The cdf of a geometric distributions is given by
F(y)  P(Y  y)  P(Y  1)  P(Y  2)  P(Y  3)  ...  P(Y  y)
p 1 q y
 p  pq  pq 2  ..... pq y 1 
1 q y
 
1 q
1 q
Let Y~ Geo(p), then   E (Y )  and Var ( X )   2  2 Show?
p p
Example 1 A sharpshooter normally hits the target 70% of the time.

a) Find the probability that her first hit is on the second shot
b) Find the mean and standard deviation of the number of shots required to realize the
1st hit
Solution
Let X be the random variable ‘the number of shoots required to realize the 1st hit’
x ~ Geo(0.7) and P( X  x)  0.71  0.7 , x  1, 2 , 3, ....
x 1
a) P( X  2)  p1     0.70.3  0.21
1 1 1- p 1  0.7
b)     1.428571 and     0.78
 0.7 p 0.7
Example 2
The State Department is trying to identify an individual who speaks Farsi to fill a foreign
embassy position. They have determined that 4% of the applicant pool are fluent in Farsi.
a) If applicants are contacted randomly, how many individuals can they expect to
interview in order to find one who is fluent in Farsi?
b) What is the probability that they will have to interview more than 25 until they find
one who speaks Farsi?
Solution
1 1
a)     25
 0.04
b) P( X  25)  1  q 25  1  0.96  0.6396  P( X  25)  1  P( X  25)  0.3604
25
Example 3
From past experience it is known that 3% of accounts in a large accounting population are in
error. What is the probability that 5 accounts are audited before an account in error is found?
What is the probability that the first account in error occurs in the first five accounts audited?
Solution
 
P(Y  5)  1  F (5)  1  1  0.975  0.975  0.8587 P(Y  5)  1  0.975  0.1413
Exercise 2.3
1. Over a very long period of time, it has been noted that on Friday’s 25% of the customers
at the drive-in window at the bank make deposits. What is the probability that it takes 4
customers at the drive-in window before the first one makes a deposit.
2. It is estimated that 45% of people in Fast-Food restaurants order a diet drink with their
lunch. Find the probability that the fourth person orders a diet drink. Also find the
probability that the first diet drinker of th e day occurs before the 5th person.
3. What is the probability of rolling a sum of seven in fewer than three rolls of a pair of
dice? Hint (The random variable, X, is the number of rolls before a sum of 7.)
4. In New York City at rush hour, the chance that a taxicab passes someone and is
available is 15%. a) How many cabs can you expect to pass you for you to find one that
is free and b) what is the probability that more than 10 cabs pass you before you find
one that is free.
5. An urn contains N white and M black balls. Balls are randomly selected, one at a time,
until a black ball is obtained. If we assume that each selected ball is replaced before the
next one is drawn, what is;
a) the probability that exactly n draws are needed?
b) the probability that at least k draws are needed?
c) the expected value and Variance of the number of balls drawn?
6. In a gambling game a player tosses a coin until a head appears. He then receives $2n ,
where n is the number of tosses.
a) What is the probability that the player receives $8.00 in one play of the game?
b) If the player must pay $5.00 to play, what is the win/loss per game?
7. An oil prospector will drill a succession of holes in a given area to find a productive
well. The probability of success is 0.2.
a) What is the probability that the 3rd hole drilled is the first to yield a productive well?
b) If the prospector can afford to drill at most 10 well, what is the probability that he will
fail to find a productive well?
8. A well-travelled highway has itstraffic lights green for 82% of the time. If a person
travelling the road goes through 8 traffic intersections, complete the chart to find a) the
probability that the first red light occur on the nth traffic light and b) the cumulative
probability that the person will hit the red light on or before the nth traffic light.
9. An oil prospector will drill a succession of holes in a given area to find a productive
well. The probability of success is 0.2.
a) What is the probability that the 3rd hole drilled is the first to yield a productive well?
b) If the prospector can afford to drill at most 10 well, what is the probability that he will
fail to find a productive well?
2.1.5 The negative binomial distribution

Suppose a Bernoulli trial is performed until the tth success is realized. Then the random
variable “the number of trials until the tth success is realized” has a negative binomial
distribution
Definition: A random variable X has the negative binomial distribution, also called the Pascal
distribution, denoted X ~ NB(r, p) , if there exists an integer n  1 and a real number
p  (0 , 1) such that P( X  r  x)r  x 1Cx  p r (1  p) x  1, 2, 3,. . .
If r=1 the negative binomial distribution reduces to a geometric distribution.
2.1.6 Hyper geometric Distribution

Hyper geometric experiments occur when the trials are not independent of each other and
occur due to sampling without replacement hyper-geometric probabilities involve the
multiplication of two combinations together and then division by the total number of
combinations
Suppose we have a population of N elements that possess one of two characteristics, e.g. D of
them are defective and N  D are non defective. A sample of n elements is randomly
selected from the population. The r.v. of interest, Y, is the number of defective elements in
the sample.
Definition: A r.v. Y is said to have a hyper geometric probability distribution if and only if
D C y  N  D Cn  y
P(Y  y)  for y  1, 2, 3,. . . n y  n and n - y  N  D
N Cn
Theorem: If Y is a r.v with a hyper geometric distribution, then;
nD nD  D  N  n 
  E (Y )  and var ( X )   2  1   
N N  N  N  1 
Example 1
Boxes contain 2000 items of which 10% are defective. Find the probability that no more than
2 defectives will be obtained in a sample of size 10 drawn Without Replacement
Solution
Let Y be the number of defectives
C C C C C
P(Y  2)  P(Y  0)  P(Y  1)  P(Y  2)  180 10  20 1 180 9  20 2 180 8
200C10 200C10 200C10
 0.3398  0.3974  0.1975  0.9347  P(Y  2)  0.0653
Example 2
How many ways can 3 men and 4 women be selected from a group of 7 men and 10 women?
Solution
C2 10 C4
7 7350
The answer is = = 0.3779 (approx)
17 C7 19448
Note that the sum of the numbers in the numerator are the numbers used in the combination
in the denominator.
This can be extended to more than two groups and called an extended hypergeometric
problem.
Exercise 2.4
1. A bottle contains 4 laxative and 5 aspirin tablets. 3 tablets are drawn at random from the
bottle. Find the probability that; a) exactly one, b) at most 1 c) at least 2 are laxative
tablet.
2. Want is the probability of getting at most 2 diamonds in the5 selected without
replacement from a well shuffled deck?
3. A massager has to deliver 10 out of 16 letters to computing department the rest to
statistics department. She mixed up the letters and delivered 10 letters at random to
computing department. What is the probability that, only 6 letters for computing
department actually got there?
4. In a class there are 20 students. 6 are compulsive smokers and they always keep
cigarette in their lockers. One day prefects checked at random on 10 lockers. What is the
probability that they find cigarette in at most 2 lockers?
5. A box holds 8 green, 4 white and 8 red beads. 6 beads are drawn at random without
replacement from the box. What is the probability that 3 red, 2 green and 1 white beads
are drawn?
2.2 Continuous Distribution

2.2.1 Uniform (Rectangular) Distribution
A uniform density function is a density function that is constant, (Ie all the values are equally
likely outcomes over the domain). Often referred as the Rectangular distribution because the graph of
the pdf has the form of a rectangle, making it the simplest kind of density function. The uniform
distribution lies between two values on the x-axis. The total area is equal to 1.0 or 100%
within the rectangle
Definition: A random variable X has a
uniform distribution over the range [a, b] If
 1 , a xb
f(x)   b  a We denote this
 0 , elsewhere
distribution by X ~ U(a, b) where: a and b
are the smallest and largest value
respectively the variable can assume..
The mean and Variance of X are given by  

ab
and  2 
b  a  respectively. 2
2 12
 0 xa
1 x xa 
The cdf F(x) is given by F(x) 
ba  dt 
ba
 F(x)   bxaa , a  x  b
 1 xb
a

Example Prof Hinga travels always by plane. From past experience he feels that
take off time is uniformly distributed between 80 and 120 minutes after check in.
Time is precious, but we do not know yet how precious it really is. We will only know when we are no longer able to take
advantage of it…
determine the probability that: a) he waits for more than 15 minutes for take-off after
check in. b) the waiting time will be between 1.5 standard deviation from the mean,
Solution
 401 , 80  x  120
X ~ U(80,120)  f(x)  
 0, elsewhere
P(X  105)  1 - P(X  105)  1 - 1054080  83
3 ba
dx  401 x4  1.51  0 
 1.5 40
P(  1.5  x    1.5 )  
 1.5
1
But   
 1.5 40 40 12 12
3 3
P(  1.5  x    1.5 )  
40 12
Exercise 2.5
1. Uniform: The amount of time, in minutes, that a person must wait for a bus is
uniformly distributed between 0 and 15 minutes, inclusive. What is the probability
that a person waits fewer than 12.5 minutes? What is the probability that will be
between 0.5 standard deviation from the mean,
2. The current (in mA) measured in a piece of copper wire is known to follow a
uniform distribution over the interval [0, 25]. Write down the formula for the
probability density function f(x) of the random variable X representing the current.
Calculate the mean and variance of the distribution and find the cumulative
distribution function F(x)
3. Slater customers are charged for the amount of salad they take. Sampling
suggests that the amount of salad taken is uniformly distributed between 5 ounces
and 15 ounces. Let x = salad plate filling weight, find the expected Value and the
Variance of x. What is the probability that a customer will take between 12 and 15
ounces of salad?
4. The thickness x of a protective coating applied to a conductor designed to work in
corrosive conditions follows a uniform distribution over the interval [20, 40]
microns. Find the mean, standard deviation and cumulative distribution function
of the thickness of the protective coating. Find also the probability that the coating
is less than 35 microns thick.
5. The average number of donuts a nine-year old child eats per month is uniformly
distributed from 0.5 to 4 donuts, inclusive. Determine the probability that a
randomly selected nine-year old child eats an average of;
a) more than two donuts
b) more than two donuts given that his or her amount is more than 1.5 donuts.
6. Starting at 5 pm every half hour there is a flight from Nairobi to Mombasa.
Suppose that none of these plane tickets are completely sold out and they always
have room for passagers. A person who wants to fly to Mombasa arrives at the
airport at a random time between 8.45 AM and (.45 AM. Determine the
probability that he waits for
a) At most 10 minutes b) At least 15 minutes
2.2.2 Exponential Distribution

The exponential distribution is often concerned with the amount of time until some
specific event occurs. For example, the amount of time (beginning now) until an
earthquake occurs has an exponential distribution. Other examples include the length,
in minutes, of long distance business telephone calls, and the amount of time, in
months, a car battery lasts. It can be shown, too, that the amount of change that you
have in your pocket or purse follows an exponential distribution. Values for an
advantage of it…
exponential random variable occur in the following way. There are fewer large values
and more small values. For example, the amount of money customers spend in one
trip to the supermarket follows an exponential distribution. There are more people that
spend less money and fewer people that spend large amounts of money.
The exponential distribution is widely used in the field of reliability. Reliability deals
with the amount of time a product lasts
In brief this distribution is commonly used to model waiting times
betweenoccurrences of rare events, lifetimes of electrical or mechanical devices
Definition: A RV X is said to have an exponential distribution with parameter   0 if

e  x for x  0 and   0
the pdf of X is: f(x)   we abbreviate this as X ~ exp( )
 0 otherwise
 is called the rate parameter
1 1
The mean and variance of this distribution are   and  2  2 respectively
 
1  e  x for x  0
The cumulative distribution function is F(x) is given by F(x)  
 0 otherwise
0.01e 0.01t , T  0
Example Torch batteries have a lifespan T years with pdf f(t)   .
 0 otherwise
Determine the probability that the battery; a) Falls before 25 hours. b) life is
between 35 and 50 hours. c) life exceeds 120 hours. d) life exceeds the mean
lifespan.
Solution
25
a) P(T  25)  F(25)   0.01e 0.01t dt  1  e 0.01( 25)  0.2212
0
50

50
b) P(35  T  50)  0.01e 0.01t dt  e 0.01t  e 0.35  e 0.50  0.0982
35 35
 
c) P(T  120)   0.01e 0.01t dt  e 0.01t  e 1.2  0  0.3012
120 120
1  
d)    100  P(T  100)   0.01e 0.01t dt  e 0.01t  e 1  0.3679
0.01 100 100
Exercise 2.6:
1. Jobs are sent to a printer at an average of 3 jobs per hour.
a) What is the expected time between jobs?
b) What is the probability thatthe next job is sent within 5 minutes?
2. The time required to repair a machine is an exponential random variable
with rate λ= 0.5 downs/hour
a) what is the probability that a repair time exceeds 2 hours?
b) what is the probability that the repair time will take at least 4 hours
given that the repair man has been working on the machine for 3
hours?
3. Buses arrive to a bus stop according to an exponential distribution with
rate λ= 4 busses/hour. If you arrived at 8:00 am to the bus stop,
a) what is the expected time of the next bus?
b) Assume you asked one of the people waiting for the bus about the
arrival time of the last bus and he told you that the last bus left at
7:40 am. What is the expected time of the next bus?
advantage of it…
4. Break downs occur on an old car with rate λ= 5 break-downs/month. The

owner of the car is planning to have a trip on his car for 4 days.
a) What is the probability that he will return home safely on his car.
b) If the car broke down the second day of the trip and the car was fixed,
what is the probability that he doesn’t return home safely on his car.
5. Suppose that the amount of time one spends in a bank is exponentially distributed
with mean 10 minutes. What is the probability that a customer will spend more
than 15 minutes in the bank? What is the probability that a customer will spend
more than 15 minutes in the bank given that he is still in the bank after 10
minutes?
6. Suppose the lifespan in hundreds of hours, T, of a light bulb of a home lamp is
exponentially distributed with lambda = 0.2. compute the probability that the light
bulb will last more than 700 hours Also, the probability that the light bulb will last
more than 900 hours
7. Let X = amount of time (in minutes) a postal clerk spends with his/her customer.
The time is known to have an exponential distribution with the average amount of
time equal to 4 minutes.
a) Find the probability that a clerk spends four to five minutes with a randomly
selected customer.
b) Half of all customers are finished within how long? (Find median)
c) Which is larger, the mean or the median?
8. The length of life of a certain type of electronic tube is exponentially distributed
with a mean life of 500 hours. Find the probability that;
a) a tube will last more than 800 hours
b) a tube will fail within the first 200 hours.
c) the length of life of a tube will be between 400 and 700 hours.
9. On the average, a certain computer part lasts 10 years. The length of time the
computer part lasts is exponentially distributed.
a) What is the probability that a computer part lasts more than 7 years?
b) On the average, how long would 5 computer parts last if they are used one
after another?
c) Eighty percent of computer parts last at most how long?
d) What is the probability that a computer part lasts between 9 and 11 years?
10. Suppose that the length of a phone call, in minutes, is an exponential random
variable with decay parameter = 1/12 . If another person arrives at a public
telephone just before you, find the probability that you will have to wait more than
5 minutes. Let X = the length of a phone call, in minutes. What is median mean
and standard deviation of X?
2.2.3 Gamma Distribution

Gamma Function

Let   0 we define ( )   x 1e  x dx called the gamma function with parameter
0
Theorem
i) ( )  (  1)(  1) if   1 ii) ( )  (  1)! for   Z 

iii) ( )  2 u 2 1eu du and iv) ( 12 )  
2
0
Proof


( )   x 1e x dx   x 1e x
0

0

 (  1)  x 2e x dx  (  1)(  1)
0
For   1 , (1)  0

 
e x dx   e x

0  1 Now suppose it holds for   k ie
(k )  (k  1)! for k  Z  then (k  1)  k (k  1)! k! for k  Z  . Thus if the results holds for
  k then they must also hold for   k  1 . But the results are true for   1
Therefore the results are true for any   Z 
  
from ( )   x 1e  x dx put x = u 2  dx  2udu  ( )   u 2 1e u 2udu  2 u 2 1e u du
2 2
0 0 0
Gamma Distribution
The Gamma  ,   distribution models the time required for  events to occur, given that
the events occur randomly in a Poisson process with a mean time between events of  . For
example, an insurance company observes that large commercial fire claims occur randomly
in time with a mean of 0.7 years between claims. Not only in real life, the Gamma
distribution is also wildly used in many scientific areas, like Reliability Assessment, Queuing
Theory, Computer Evaluations, or biological studies. In a nut shell, this distribution is used
to model total waiting time of a procedure that consists of  independent stages, each stage
with a waiting time having a distribution Exp(  ) .Then the total time has a Gamma
distribution with parameters  and  .
Definition A random variable X has gamma density if it pdf is given by

x
x 1e  ,
f(x)  x  0 and f(x)  0 otherwise. Abbreviated as X ~ Gamma  ,  
( )  
The parameter  is called the shape parameter while  is called the rate or scale parameter
( ) is the Gamma function.
Some examples of gamma distributions are plotted below. Notice that the modes shift to the
right as the ratio of  increases.
Remarks
a) If   1 then we have the standard gamma distribution.
b) If   1 then we have exponential density function.

1 x
 0
The cdf, F(x) is of the form F(x)  t  1e  t dt and its computation is not trivial.
( ) 
Theorem: If X has a gamma distribution with parameters  and  , then
E(X)     and Var (X)   2   2
Definition: Let  be a positive integer. A random variable X is said to have a chi-square
distribution with  degree of freedom if and only if is a gamma-distributed random variable
with parameters   2 and   2
Theorem: If X is a chi-square random variable with  degrees of freedom, then
E(X)     and Var (X)   2  2
Example 1 Suppose the reaction time of a randomly selected individual to a certain

stimulus has a standard gamma distribution with   2 sec. Find the probability that reaction
time will be (a) between 3 and 5 seconds (b) greater than 4 seconds
Solution
(2)  1! . Therefore f(x)  xe  x , x  0 and f(x)  0 elsewhere
5
P(3  X  5)   xe  x dx   ( x  1)e  x  4e 3  6e 5  0.1587
5
3 3
P( X  4)  1  P( X  4)  1   xe dx  1   ( x  1)e  x   1  1  5e  4  0.09158
0
4

x
0

4
 
Example 2 Suppose the survival time X in weeks of a randomly selected male mouse
exposed to 240 rads of gamma radiation has a gamma distribution with   8 and   15 .
a) Find the expected value and the standard deviation of the survival time.
b) What is the probability that a mouse survives (i) between 60 and 120 weeks.(ii) at least 30
weeks
Solution
X ~ 8 ,15  f(x)  8
1 x
x 7 e 15 , x  0 and f(x)  0 otherwise
15 (8)
a) E(X)   = (8)(15) = 120 weeks   8(15) 2  30 2  42.4264 weeks
120 1 8 1
P(60  X  120)   7  15
4 (8) y e dy
7 y
x
b) x e dx 
60 158  (8)
8
e y

  y  7 y  42 y  210 y  840 y  2520 y  5040 y  5040
7 6 5 4 3
7!
2

4
4 8
261104e  6805296e
  0.4959
7!
 1  1
PX  30)   x 7 e 15 dx  
x
c) y 7 e  y dy
30 15  (8)
8 2 7!

e y

  y  7 y  42 y  210 y  840 y  2520 y  5040 y  5040
7 6 5 4 3
7!
2

2
2
37200e  0
  0.9989
7!
Question The time between failures of a laser machine is exponentially distributed with a
mean of 25,000 hours. What is the expected time until the second failure? What is the
probability that the time until the third failure exceeds 50,000 hours?
2.2.4 Beta Distribution

Beta Function
We define a beat function as B  ,   as B  ,     x 1 1  x 
1  1
dx with parameters
0
  0 and   0
Theorem:
i) B  ,    B  ,   ie by putting y  1  x in the function
 u 1
ii) B  ,    
x u
du ie by putting u  or x  in the function
0 1  u  1
1 x 1 u

iii) B  ,    2 cos 2 1 t sin 2  1 tdt ie by putting x  cos 2 t in the function

2
0
( )(  )
iv) B  ,   
(   )
( 12 )( 12 )
v) ( 12 )   use B  12 , 12    ( 12 ) but
2
(1)
B  12 , 12   2 cos
 
tdt  2 dt  
2 12 1 2 12 1
 ( 12 )  
2 2
t sin
0 0
 1  
Note ( 12 )   x e dx  2 e u du   . Simply
2 x
 e  x dx  12 
2 2
0 0 0
Beta Distribution
Definition: A random variable X is said to have a standard beta distribution with parameters
 and  if it’s probability density function is given by
x 1 1  x 
 1
f(x)  ,0  x  1 and f(x) = 0 elsewhere we denote this as X ~ Beta  ,  
B( ,  )
Theorem: If X has a standard beta distribution with parameters  and  , then
 
E(X)    and. Var (X)   2 
      1   2
3. MOMENTS AND MOMENT-GENERATING FUNCTIONS

Definition: The kth moment of a r.v. X taken about zero, or about origin is defined to be
 
E X k and denoted by  k/ .
Definition: The kth moment of a r.v. X taken about its mean, or the kth central moment of X,
is defined to be E ( X   ) k and denoted by  k .
Definition: The moment-generating function (mgf), for a r.v. X denoted as M x (t ) or simply
M(t ) is given by M (t )  E etx  
We say that an mgf for X exists if there is b > 0 such that M (t )   for t  b .
tx (tx ) 2 (tx )3 (tx ) 4

Notice that etx  1      ..... therefore
1! 2! 3! 4!
 tx (tx ) 2 (tx )3 (tx ) 4 
M (t )  E etx     1      ....  P( X  x)
all x  1! 2! 3! 4! 
t2 t3
  P( X  x)  t  xP ( X  x)   x P( X  x)   x 3 P( X  x)  ....
2
all x all x 2! all x 3! all x

t2 t3
 1  tE ( X )  E ( X 2 )  E ( X 3 )  ... ie a function of all moments about the origin
2! 3!
d k M (t )
Theorem: If M (t ) exists, then for any k  N k
 k/  E ( X k )
dt t 0
Proof
t2 t3
M (t )  1  t1/  2/  3/  ...
2! 3!
2
2t 3t /
M ' (t )  1/   2/  3  ...  M ' (0)  1/  E ( X )
2! 3!
2t 3t 2 /
M ' ' (t )   2/  3/   4  ...  M ' ' (0)   2/  E ( X 2 )
2! 3!
Remark: The mgf of a particular distribution is unique and we can recognize the pdf if we
are given the mgf.
Example 1
The mgf of a r.v Y is given by M (t )  16 et  13 e2t  12 e3t Find the mean and variance of Y
Solution

E (Y )  M ' (0)  16 et  23 e2t  32 e3t  16  23  32  73  t 0
E (Y )  M ' ' (0) 

2
 1
6
e  e  e
t 4
3
2t 9
2
3t
 t 0
 16  43  92  6
Var (Y )  E (Y 2 )   2  6   73   95
2
Example 2 Find the mgf of a r.v X ~ Po ( )

Solution
 
M (t )  E e   e

  xe 
e  t x
 e ee  e e 1
e 

t t
tx tx
x 0 x! x 0 x!
Now E ( X )  M ' (0)  et e e 1 
t
t 0
 1  5  for x  0 ,1, 2 , 3.....

 x
Example 3 Find the mgf of a r.v X whose pmf is given by f(x)   6 6


 0 elsewhere
hence obtain the mean and variance of X
Solution
     
 
. M (t )  E etx   etx 16  56    16 56 et  65 t 
1
x 1 1
 6  5et
x
x 0 x 0 1  6 e 6  5e t

M ' (t )  5et 6  5et 
2
and 
M ' ' (t )  5et 6  5et 
2

 50e2t 6  5et 3
  )  5e 6  5e    
2 t 2 3
 E ( X )  5et 6  5et  5 and E (Y 2 t
 50e2t 6  5et t 0  55
t 0
Var ( X )  E ( X 2 )   2  55  52  30
Exercise 3.1
1) The mgf of a r.v Y is given by; a) M (t )  e2t 3t b) M (t )  exp 12  2t 2  t Find the
2
mean and variance of Y

2) A r.v X has a gamma distribution with parameters  and  , Find the mgf of X hence
obtain the mean and variance of X
3.1 The Mgf of a Sum of Independent Random Variables

The mgf of the sum of n independent random variable is the product of their individual mgf’s
The mean (variance) of the sum of n independent random variable is the sum of their
individual means (variances).

The mgf of a r.v X about X  a is given by M x , a (t )  E et  xa   e at E etx  e at M x (t )   
4 NORMAL DISTRIBUTION
4.1 Introduction
The normal, or Gaussian, distribution is one of the most important distributions in probability
theory. It is widely used in statistical inference. One reason for this is that sums of random
variables often approximately follow a normal distribution.
Definition A r.v X has a normal distribution with parameters  and  2 , abbreviated
 
X ~ N  ,  2 if it has probability density function
f(x)   1
2

exp  12    for    x  
x 2
 and   0
Where  is the mean and  is the standard deviation.
4.1.1 Properties of normal distribution
1) The normal distribution curve is bell-shaped and symmetric, about the mean
2) The curve is asymptotic to the horizontal axis at the extremes.
3) The highest point on the normal curve is at the mean, which is also the median and mode.
4) The mean can be any numerical value: negative, zero, or positive
5) The standard deviation determines the width of the curve: larger values result in wider,
flatter curves
6) Probabilities for the normal random variable are given by areas under the curve. The total
area under the curve is 1 (0.5 to the left of the mean and 0.5 to the right).
7) It has inflection points at    and    .
8) Empirical Rule:
a) 68.26% of values of a normal random variable are within  1 standard deviation of its
mean. ie P    X       0.6826
b) 95.44% of values of a normal random variable are within  2 standard deviation of its
mean. ie P  2  X    2   0.9544
c) 99.72% of values of a normal random variable are within  3 standard deviation of its
mean. ie P  3  X    3   0.9972
4.2 Standard Normal Probability Distribution

A random variable having a normal distribution with a mean of 0 and a variance of 1 is said
to have a standard normal probability distribution
Definition The random variable Z is said to have the standard normal distribution if
Z ~ N0 ,1 . Therefore, the density of Z, which is usually denoted  (z) is given by;
 (z) 
1

exp  12 z 2 for    z   
2
The cumulative distribution function of a standard normal random variable is denoted (z ) ,
and is given by
z
( z )    (t )dt
-
4.2.1 Computing Normal Probabilities

It is very important to understand how the standardized normal distribution works, so we will
spend some time here going over it. There is no simple analytic expression for (z ) in terms
of elementary functions. but the values of (z ) has been exhaustively tabulated. This greatly
simplifies the task of computing normal probabilities..
Table 1 below reports the cumulative normal probabilities for normally distributed variables
in standardized form (i.e. Z-scores). That is, this table reports P(Z  z) = ( z) ). For a given
value of Z, the table reports what proportion of the distribution lies below that value. For
example, P(Z  0) = (0)  0.5 ; half the area of the standardized normal curve lies to the left
of Z  0 .
Theorem: It may be useful to keep in mind that

i) P(Z  z) = 1 - ( z) complementary law
ii) P(Z   z) = P(Z  z)  1 - ( z) ie due to symmetry

 ( z)  ( z)  1 Since P(Z  z)  P(Z  z)  1
iii) P(a  z  b) = (b) - (a)
iv) P(-a  z  a) = 2(a)  1since
P(-a  z  a) = (a)  (a) = (a)  1  (a)  2 (a)  1
v) If we now make (a) the subject, then (a) = 12 1  P(-a  z  a)
Table 1
Example 1
Given Z ~ N0 ,1 , find; d) P(-0.696  z  1.865)
a) P(Z  z) if z = 1.65, -1.65, 1.0, -1.0 e) P(-2.345  z  1.65)
b) P(Z  z) for z = 1.02, -1.65 f) P( z  1.43)
c) P(0.365  z  1.75)
Solution
a) Look up and report the value for (z ) from the standard normal probabilities table
P(Z  1.65) = (1.65)  0.9505 (1.65)  0.0495 (1.0)  0.8413 (1.0)  0.1587
b) P(Z  z)  ( z) Thus P(Z  1.02)  (1.02)  0.1515 P(Z  -1.65)  (1.65)  0.9505
c) P(0.365  z  1.75)  (1.75) - (0.365)  0.9599 - 0.6350  0.3249
d) P(-0.696  z  1.865)  (1.865) - (-0.696)  0.9689 - 0.2432  0.3249  0.7257
e) P(-2.345  z  1.65)  (1.65) - (-2.345)  0.0505  0.0095  0.0410
f) P( z  1.43)  P(-1.43  z  1.43)  2(1.43) -1  2(0.9236)  1  0.8472
Example 2 If Z ~ N0 ,1 , find the value of t for which;

a) P(Z  t) = = 0 .6026, 0.9750, 0.3446 c) P(-0.28  z  t )  0.2665
b) P(Z  t) = 0.4026, 0.7265, 0.5446 d) P(-t  z  t )  0.9972 , 0.9505 , 0.9750
Solution
Here we find the probability value in Table I, and report the corresponding value for Z.
a) ( t) = 0 .6026  t  0.26 ( t) = 0 .950  t  1.96 ( t) = 0 .3446  t  0.40
b) P(Z  t)  0.4026  ( t) = 0.5974  t  0.25
P(Z  t)  0.7265  ( t) = 0.2735  t  - 0.60
P(Z  t)  0.5446  ( t) = 0.4554  t  - 0.11
c) P(-0.28  z  t )  (t) - (-0.28) = 0.2665  (t)  0.3897  0.2665  t = 0.40
d) P(-t  z  t )  2 (t) -1 = 0.9972  (t)  0.9986  t = 2.99
P(-t  z  t )  2 (t) -1 = 0.9505  (t)  0.9753  t = 1.96
P(-t  z  t )  2 (t) -1 = 0.9750  (t)  0.9875  t = 2.24
Exercise 4.1
1...Given Z ~ N0 ,1 , find; 2...If Z ~ N0 ,1 , find the value of z for
a) P(Z  z) if which;
z = 1.95, -1.89, 1.074, -1.53 a) P(Z  a) = = 0 .973, 0.6693, 0.4634
b) P(Z  z) for z = 1.72, -1.15 b) P(Z  a) = 0.3719, 0 .9545, 0 .7546
c) P(0  z  1.05) c) P(-1.21  z  t )  0.6965
d) P(-1.396  z  1.125) d) P( z  t )  0.9544 , 0.9905 , 0.3750
e) P(-1.96  z  1.65)
f) P( z  2.33)
4.3 The General Normal Density

Consider Z ~ N0 ,1 and let X    Z for   0 . Then X ~ N ,  2  But we know that
f (x)  1   from which the claim follows. Conversely, if X ~ N , , then

X-

2
Z X-
 ~ N0 ,1 . It is also easily shown that the cumulative distribution function satisfies
F(x)   X-   
and so the cumulative probabilities for any normal random variable can be calculated using
the tables for the standard normal distribution..
Definition A variable X is said to be standardized if it has been adjusted (or transformed)

such that its mean equals 0 and its standard deviation equals 1. Standardization can be
accomplished using the formula for a z-score: Z  X-  ~ N0 ,1 . The z-score represents the
number of standard deviations that a data value is away fromthe m ean.
 
Let X ~ N  ,  2 then P(a  X  b)  P a -
 Z  b- 

     where
b- 

a -

X- 
Z  ~ N (0 ,1)
Example 1 A r.v X ~ N50 , 25 compute P(45  z  60)

Solution
  50 and   5  Z  x -550 ~ N (0 ,1)
P(45  X  60)  P 455-50  Z  605-50   (2)  (1)  0.9772 - 0.1587  0.8185
Example 2 Suppose X ~ N30 , 16 . Find; a) P( X  40) b) P( X  21) c) P(30  X  35)

Solution
X ~ N30 , 16  Z  X430 ~ N 0, 1
a) P( X  40)  PZ  404-30   (2.5)  0.9938
b) P( X  21)  PZ  21430   PZ  2.25  PZ  2.25  (2.25)  0.9878
c) P(30  X  35)  P30430  Z  35430   P(0  Z  1.25)  0.8944  0.5  0.3944
Example 3
The top 5% of applicants (as measured by GRE scores) will receive scholarships. If
GRE ~ N(500,100 2 ) , how high does your GRE score have to be to qualify for a scholarship?
Solution
Let X  GRE . We want to find x such that P(X  x) = 0.05 this is too hard to solve as it
stands - so instead, compute Z  X100 500
~ N 0, 1 and find z for the problem,
P(Z  z)  1 - ( z) = 0.05  ( z)  0.95  z  1.645
To find the equivalent x, compute X    Z  x  500  100(1.645)  66.5
Thus, your GRE score needs to be 665 or higher to qualify for a scholarship.
Example 4 Family income is believed to be normally distributed with a mean of $25,000

and a standard deviation on $10,000. If the poverty level is $10,000, what percentage of the
population lives in poverty? A new tax law is expected to benefit “middle income” families,
those with incomes between $20,000 and $30,000. What percentage of the population will
benefit from the law?
Solution
Let X = Family income. We want to find P(X ≤ $10,000)., so
 
X ~ N 25000 , 100002  Z  X10000
 25000
~ N 0, 1
P( X  10,000)  PZ  1.5  (1.5)  0.0668 .
Hence, a slightly below 7% of the population lives in poverty.
P(20,000  X  30,000)  P 0.5  Z  0.5  2(0.5) 1  2  0.6915 -1  0.383

Thus, about 38% of the taxpayers will benefit from the new law.
Exercise 4.2
1) Suppose X ~ N130 , 25 . Find; a) P( X  140) b) P( X  120) c) P(130  X  135)
2) The random variable X is normally distributed with mean 500 and standard deviation 100.
Find; (i) P( X  400) , (ii) P( X  620) (iii) the 90th percentile (iv) the lower and upper
quartiles. Use graphs with labels to illustrate your answers.
3) A radar unit is used to measure speeds of cars on a motorway. The speeds are normally
distributed with a mean of 90 km/hr and a standard deviation of 10 km/hr. What is the
probability that a car picked at random is travelling at more than 100 km/hr?
4) For a certain type of computers, the length of time bewteen charges of the battery is
normally distributed with a mean of 50 hours and a standard deviation of 15 hours. John
owns one of these computers and wants to know the probability that the length of time
will be between 50 and 70 hours
5) Entry to a certain University is determined by a national test. The scores on this test are
normally distributed with a mean of 500 and a standard deviation of 100. Tom wants to
be admitted to this university and he knows that he must score better than at least 70% of
the students who took the test. Tom takes the test and scores 585. Will he be admitted to
this university?
6) A large group of students took a test in Physics and the final grades have a mean of 70
and a standard deviation of 10. If we can approximate the distribution of these grades by a
normal distribution, what percent of the student; (a) scored higher than 80? (b) should
pass the test (grades≥60)? (c) should fail the test (grades<60)?
7) A machine produces bolts which are N(4 0.09) where measurements are in cm. Bolts are
measured accurately and any bolt smaller than 3.5 cm or larger than 4.4 cm is rejected.
Out of 500 bolts how many would be accepted? Ans 430
8) Suppose IQ ~ N (100, 22.5).a woman wants to form an Egghead society which only
admits people with the top 1% IQ score. What should she have to set the cut-off in the
test to allow this to happen? Ans 134.9
9) A manufacturer does not know the mean and standard deviation of ball bearing he is
producing. However a sieving system rejects all the bearings larger than 2.4 cm and those
under 1.8 cm in diameter. Out of 1,000 ball bearings, 8% are rejected as too small and
5.5% as too big. What is the mean and standard deviation of the ball bearings produced?
Ans mean=2.08 sigma=0.2
4.4 Normal Approximation to the Binomial Distribution.

4.4.1 Introduction
Suppose a fair coin is tossed 10 times, what is the probability of observing: a) exactly 4 heads
b) at most 4 heads?
Solution
Let X be the r.v the number of heads observed then X ~ Bin 10 , 0.5
 P( X  x)10Cx  0.5 0.5 10Cx  12  for x = 0,1,2,. . . .,10
x 10 x 10
a)  P( X  4)10 C4   12  
105
 0.2051
10
512
b) P( X  4)  10 C0 10C1 10C2 10C3 10C4 0.5 
193
 0.3770
10
512
4.4.2 Normal approximation:

Many interesting problems can be addressed via the binomial distribution. However, for large
n, it is sometimes difficult to directly compute probabilities for a binomial (n, p) random
variable, X. Eg: Compute P( X  12) for 25 tosses of a fair coin. Direct calculations can get
cumbersome very quickly. . Fortunately, as n becomes large, the binomial distribution
becomes more and more symmetric, and begins to converge to a normal distribution. That is,
for a large enough n, a binomial variable X is approximately ∼ N(np, npq). Hence, the
normal distribution can be used to approximate the binomial distribution.
To get a feel for why this might work, let us draw the probability histogram for 10 tosses of a
fair coin. The histogram looks bell-shaped, as long as the number of trials is not too small
In general, the distribution of a binomial random variable may be accurately approximated by

that of a normal random variable, as long as np  5 and nq  5 , and assuming that a
.continuity correction is made to account for the fact that we are using a continuous
distribution (the normal) to approximate a discrete one (the binomial).
In approximating the distribution of a binomial random variable X, we will use the normal
distribution with mean   np and variance  2  npq where q  1  p . Why are these
reasonable choices of μ,  2 ?
4.4.3 Continuity Correction

In the binomial, P( X  a)  P( X  a  1)  1 whenever a is an integer. But if we sum the
area under the normal curve corresponding to P( X  a)  P( X  a  1) , this area does not
sum to 1 because the area from a to (a + 1) is missing.
The usual way to solve this problem is to associate 1/2 of the interval from a to a + 1 with
each adjacent integer. The continuous approximation to the probability P( X  a) would thus
be P( X  a  12 ) , while the continuous approximation to P( X  a  1) would be
P( X  a  12 ) . This adjustment is called a continuity correction. More specifically,
P( X  x)  P( X  x  1)  P( X  x  0.5)  P z  x  0npq
    
 
.5  np

 
Binomialdistribution Normal approximation
 
Standardd Normal approx
P( X  x)  P( X  x  1)  P( X  x  0.5)  P z  x 0npq
.5 np
    
  
 
a  0.5 np
P(a  X  b)  P(a  1  X  b  1)   P(a  0.5  X  b  0.5)  P  z  b  0npq
.5 np
    npq
 
P( X  x)  P( x  1  X  x  1)  P( x  0.5  X  x  0.5)  P x 0npq

.5  np
 z  x  0npq
.5 np
     
NB: For the binomial distribution, the values to the right of each = sign are primarily
included for illustrative purposes. The equalities which hold in the binomial distribution do
not hold in the normal distribution, because there is a gap between consecutive values of a.
The normal approximation deals with this by “splitting” the difference.
For example, in the binomial, P( X  6)  P( X  7) since 6 is the next possible value of X

that is less than 7. In the normal, we approximate this by finding P( X  6.5) . And, in the
binomial, P( X  6)  P( X  5) , because 6 is the next value of X that is greater than 5. In the
normal, we approximate this by finding P( X  5.5)
Returning to the case of coin tossing. Suppose we wish to find P( X  4) , the probability that
the binomial r.v is less than or equal to 4. In the diagram above, the bars represent the
binomial distribution with n = 10, p = 0.5. The superimposed curve is a normal density f(x).
The mean of the normal is   np  5 and the standard deviation is
  10(0.5)(0.5)  1.58 Using the normal approximation, we need to calculate the
probability that our normal r.v is less than or equals to 4.5. ie
P( X  4)  P( X  4.5)  PZ  41.5.585   (0.3162)  0.3759 which is very close to the
   
Binomial Normal std normal
actual answer of 0.377
Example 1 Suppose 50% of the population approves of the job the governor is doing, and
that 20 individuals are drawn at random from the population. Solve the following, using the
normal approximation to the binomial. What is the probability that;
a) exactly 7 people will support the governor?
b) at least 7 people will support the governor?
c) more than 11 people will support the governor?
d) 11 or fewer will support the governor?
Solution
Note that n  20 , p  0.5    np  10 and   npq  5 Since np  5 and nq  5 , it is
probably safe to assume that X ~ N (10 , 5)
        
 P
a) P( X  7)  P(6.5  X  7.5)  P 6.5510  Z  7.5510    1 Z
.565 
 1
.118



std normal
 (1.118)  (1.565)  0.1318 - 0.0588  0.0730

b) P( X  7)  P( X  7.5)  PZ  1.118  0.1318
    
c) P( X  11)  P( X  11.5)  PZ  0.6708  1  0.6708  1  0.7488  0.2512

   
d) P( X  11)  P( X  11.5)  PZ  0.6708  0.6708  0.7488

   
Example 2
In each of 25 races, the Democrats have a 60% chance of winning. What are the odds that the
Democrats will win 19 or more races? Use the normal approximation to the binomial
Solution
Note that n  25 , p  0.6    np  15 and   npq  6 Since np  5 and nq  5 , it is
probably safe to assume that X ~ N (15 , 6). .
Using the normal approximation to the binomial,
P( X  19)  P( X  18.5)  PZ  1.4289  1  1.4289  1  0.9235  0.0765
   
Hence, Democrats have a little less than an 8% chance of winning 19 or more races.
Example 3 Tomorrow morning Iberia flight to Madrid can seat 370 passengers. From past
experience, Iberia knows that the probability is 0.90 that a given ticket-holder will show up
for the flight. They have sold 400 tickets, deliberately overbooking the flight. How confident
can Iberia be that no passenger will need to be .bumped (denied boarding)?
Solution:
We will assume that the number (X) of passengers showing up for the flight has a binomial
distribution with mean   400  0.9  360 and standard deviation   400  0.9  0.1  6
We want P( X  370)  P( X  370.5)  PZ  1.75  0.9599 So the probability that
    
nobody gets bumped is approximately 0.9599. (Almost 96%).
Exercise 4.3 (Use the normal approximation to the binomial in the entire exercise)
1. A coin is loaded such that heads is thrice as likely as the tails. Find the probability of
observing between 4 and 7 heads inclusive with 12 tosses of the coin.
2. Based upon past experience, 40% of all customers at Miller’s Automotive Service Station
pay for their purchases with a credit card. If a random sample of 200 customers is
selected, what is the approximate probability that;
a) at least 75 pay with a credit card?
b) not more than 70 pay with a credit card?
c) between 70 and 75 customers, inclusive, pay with a credit card?
3. The probability that Ron scores a goal in any game against a tough opponent in soccer is
0.3. What is the probability that he scores 30 goals in the next 100 games which he plays?
4. Crafty Computers limited produces PCs. The probability that one of their computers has a
virus is 0.25. JKUAT ICSIT buys 300 computers from the company. What is the
probability that between 70 and 80 PCs inclusive have a virus? Would you advice the
director JKUAT ICSIT to buy Computers from this company in future?
5. For overseas flights, an airline has three different choices on its dessert menu—ice cream,
apple pie, and chocolate cake.Based on past experience the airline feels that each dessert
is equally likely to be chosen.
a) If a random sample of four passengers is selected, what is the probability that at
least two will choose ice cream for dessert
b) If a random sample of 21 passengers is selected, what is the approximate
probability that at least eight will choose ice cream for dessert?
6. In a family of 11 children, what is the probability that there will be more boys than girls?.
7. A baseball player has a long term batting average of 0.300. What is the chance he gets an
average of 0.330 or higher in his next 100 bats?
8. Suppose we draw a Simple Random Sample of 1,500 Americans and want to assess
whether the representation of blacks in the sample is accurate. We know that about 12%
of Americans are black, what is the probability that the sample contains at most170 lacks?
9. Let T be the lifetime in years of new bus engines. Suppose that T is continuous with
0 for t < 1
probability density function f(t) =  d for some constant d.
 t 3 for t > 1
a) Find the value of d and the mean and median of T.
b) Suppose that 240 new bus engines are installed at the same time, and that their
lifetimes are independent. By using a normal approximation to the binomial, find the
probability that at most 10 of the engines last for 4 years or more.
4.5 Sums of Independent Random Variables

Theorem 1 If X 1 , X 2 , .... , X n are n independent continuous r.v each with mean  and
1 n 2
variance  2   then r.v W =  i
n i 1
x has mean  and variance
n
2
 
Theorem 2 If X 1 ~ N 1 ,  1 and X 2 ~ N 2 ,  2 are 2 independent r.v, then
2
 

X1  X 2 ~ N 1  2 ,    2
1
2
2 
Proof
t1  12 t 2 12 t 2  12 t 2 22
Let the mgf of X 1 be M1 (t )  e and for X 2 be M 2 (t )  e . Let Y1  X 1  X 2 so
t1  12 t 212 t2  12 t 2 22 t ( 1  2 )  12 t 2 (12  22 )
Y1 has mgf M1 (t )  M1 (t )  M 2 (t )  e e e which is the mgf of
2 2
a normal r.v with mean 1   2 and variance  1   2

Note Y2   X 2 ~ N 2 ,  22   X1  X 2 ~ N 1  2 , 12   22 
Theorem 3 If X1 , X 2 , ....Xn are n independent r.vs and each X i ~ N i ,  i2 , then the r.v  

X1  X 2  .....  X n ~ N 1  2  .....  n ,     .....   2
1
2
2
2
n 
Proof is by induction (left as exercise to the learner)
Example If X ~ N60 ,16 and Y ~ N70 , 9 are 2 independent r.v, Find (a) P X Y  140
(b) P120  X  Y  135 (c) PY  X  7 (d) P2  Y  X  12
Solution
X  Y ~ N130 ,25 and Y  X ~ N10 ,25 therefore
a) P X  Y  140  PZ  1405130   (2)  0.9772
b) P120  X  Y  135  P1205130  Z  1355130   (1)  (2)  0.8413 - 0.0228  0.8185
c) PY  X  7  PZ  7510   PZ  0.6  PZ  0.6  (0.6)  0.7257
d) P2  Y  X  12  P 2510  Z  12510   (0.4)  (1.6)  0.6554 - 0.0548  0.6006
Exercise 4.4
1. If X ~ N65 ,28 and Y ~ N85 , 36 are 2 independent r.v, Find (a) P X Y  142
(b) P134  X  Y  166 (c) PY  X  4 (d) P12  Y  X  24
2. Each day Mr. Njoroge walks to the library bto read a newspaper. Total time spent
walking is normally distributed with mean 15 minutes and standard deviation 2 minutes.
Total time spent in the library is also normally distributed with mean 25 minutes and
standard deviation 12 minutes. Find the probability that on one day;
a) he is away from his home for more than 45 minutes.
b) he spends more time walking than in the library
4.6 Sampling Distributions

In many investigations the data of interest can take on many possible values and it is often of
interest to estimate the population mean, μ. A common estimator for μ is the sample mean x .
Consider the following set up: We observe a sample of size n from some population and
1 n
compute the mean x   xi .Since the particular individuals included in our sample are
n i 1
random, we would observe a different value of x if we repeated the procedure. That is, x is
also a random quantity. Its value is determined partly by which people are randomly chosen
to be in the sample. If we repeatedly drew samples of size n and calculated x , we could
ascertain the sampling distribution of x .
Many possible samples, many possible x ’s
mean = 1.78 mean = 1.55 mean = 1.45 mean = 1.6 mean = 1.73
0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10
0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10
0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10
0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10
We only see one!
We will have a better idea of how good our one estimate is if we have good knowledge of
how x behaves; that is, if we know the probability distribution of x .
4.6.1 Properties of the Sampling Distribution of the Sample Means (Summary)

When all of the possible sample means are computed, then the following properties are true:
1. The mean of the sample means will be the mean of the population
2. The variance of x will be the variance of the population divided by the sample size.
3. The standard deviation of the sample means (known as the standard error of the mean) will be
smaller than the population mean and will be equal to the standard deviation of the population
divided by the square root of the sample size.
4. If the population has a normal distribution, then the sample means will have a normal distribution.
5. If the population is not normally distributed, but the sample size is sufficiently large, then the
sample means will have an approximately normal distribution. Some books define sufficiently
large as at least 30 and others as at leas t25.
Example The law firm of Hoya and Associates has six partners (A, B, C, D, E, F). At their
weekly partners meeting each reported the number of hours they charged clients for their
services last week. A 24, B 26, C 28, D 26, E 24. F26 (eg, Mr. E charged 24 hrs)
If n=2, (ie two partners) are selected randomly, how many different samples are possible?
This is the combination of 6 objects taken 2 at a time. Ie there are 6 c 2  15 possible samples.
15 sample means are given below: (e.g. if the sample has A and B, sample mean is 24)
AB 25, AC 26, AD 25, AE 24, AF 25, BC 27, BD 26, BE 25, BF 26, CD 27, CE 26, CF 27,
DE 25, DF 26, EF 25 putting this in a frequency table we get
x 24 25 26 27
f 1 6 5 3
This is almost the sampling distribution of means. If we divide individual frequencies by total
frequency (ie 15) we get “relative frequency” or probability. These probabilities add up to
one, so we have a prob. distribution. The above information says that the probability that
sample mean is 24 is 1 out of 15 or 0.066667. Now draw a histogram for sampling distribution
of means and fit a normal curve on it.
Note the shape is similar to Normal distribution
The sampling distribution is simply this probability distribution defined over all possible
samples of size n from the population of size N. In the real world problems N will be large
(e.g. 200 million US population) and n will be also be large (e.g., 1000 people surveyed) and
N
c n will be astronomical number. Then the sampling distribution can only be imagined. We
have chosen a simple example of N=6, n=2 so that the entire sampling distribution can be
explicitly computed and visualized. Now the random variable is x , it is no longer just X.
Definitions
Central Limit Theorem:- Stats that as the sample size increases, the sampling distribution of
the sample means will become approximately normally distributed.
Sampling Distribution of the Sample Means:- Distribution obtained by using the means
computed from random samples of a specific size.
Sampling Error :- Difference which occurs between the sample statistic and the population
parameter due to the fact that the sample isn't a perfect representation of the population.
Standard Error or the Mean:- The standard deviation of the sampling distribution of the
sample means. It is equal to the standard deviation of the population divided by the square
root of the sample size.
4.6.2 The Mean and Standard Deviation of x

What are the mean and standard deviation of x ?
Let’s be more specific about what we mean by a sample of size n. We consider the sample to
be a collection of n independent and identically distributed (or iid )random variables
X 1 , X 2 , ... , X n with common mean  and common standard deviation  .
1 n  1 n 1 n
Thus, E ( X )  E  xi    E xi      (n ) 
1
 n i 1  n i 1 n i 1 n
1 n
 1 n n
2 
Var ( X )  Var   xi   2 Var xi   2   2  2 (n 2 ) 
1 1
 SD( X ) 
 n i 1  n i 1 n i 1 n n n
4.6.3 The Central Limit Theorem

Now we know that x has mean  and standard deviation  n
, but what is its distribution?
If X 1 , X 2 , ... , X n are normally distributed, then x is also normally distributed. Thus,
 
X i ~ N ( , 2 )  X ~ N  , n . If X 1 , X 2 , ... , X n are not normally distributed, then the
2
Central Limit Theorem tells us that x is approximately Normal.

In brief if X 1 , X 2 , ... , X n are iid random variables with mean  and finite standard deviation
 . Then for a sufficiently large n, the sampling distribution of X is approximately Normal
with mean  and variance n .
2
Remarks
 Central limit theorem involves two different distributions: the distribution of the original
population and the distribution of the sample means
x
 The formula for a z-score when working with the sample means is: z  ~ N(0 ,1)
/ n
Example Intelligence Quotient (IQ) is normally distributed with mean 110 and standard
deviation of 10. A moron is a person with IQ less than 80. Find the probability that a
randomly chosen person is a moron. Let idiot be defined as one with an IQ less than 90.
Find the probability that a randomly chosen person is an idiot. (Hint this random variable is
for a single person X)
If a sample of 25 students is available, what is the probability that the average IQ exceeds
105? What is the probability that the average IQ exceeds 115 (Hint this random variable is for
an average over 25 persons or X )
Solution
 
IQ  X ~ N 1101 , 102 , and therefore for a sample of 25 people average IQ  X ~ N1101 , 4
The probability that a randomly chosen person is a moron is given by
P( X  80)  PZ  8010
110
  (3)  0.0013
The probability that a randomly chosen person is an idiot is given by
P( X  90)  PZ  9010
110
  (2)  0.0228
The probability that the average IQ exceeds 105 is P( X  105)

The random variable under consideration here is the average. Hence, a sampling distribution
is relevant when we consider average IQ as the variable of interest, not he IQ of an individual
student, but the average over 25 students. Standard deviation of the sampling distribution =
Standard Error SE =  n  105  2 . Now
P( X  105)  P( Z  1052110)  P( Z  2.5)  (2.5) = 0.9938

dueto symmetry
We now find probability that the average IQ exceeds 115 ie

P( X  115)  P( Z  1152110)  P( Z  2.5)  (2.5) = 0.0062

dueto symmetry
Exercise 4.5
1) The annual salaries of employees in a large company are approximately normally
distributed with a mean of $50,000 and a standard deviation of $10,000. If a random
sample of 50 employees is taken, what is the probability that their average salary is;
a) less than $45,000?
b) between $45,000 and $65,000?
c) more than $70,000
2) Library usually has 13% of its books checked out. Find the probability that in a sample of
588 books greater than 14% are checked out. ANS= 0.2358
3) The length of similar components produced by a company are approximated by a normal
distribution model with a mean of 5 cm and a standard deviation of 0.02 cm.
a) If a component is chosen at random what is the probability that the length of this
component is between 4.98 and 5.02 cm?
b) what is the probability that the average length of of a sample of 25 component is
between 4.96 and 5.04 cm?
4) The length of life of an instrument produced by a machine has a normal distribution with
a mean of 12 months and standard deviation of 2 months. Find the probability that in a
random sample of 4 instrument produced by this machine, the average length of life
a) less than 10.5 months. b) between 11and 13 months.
5) The time taken to assemble a car in a certain plant is a random variable having a normal
distribution of 20 hours and a standard deviation of 2 hours. What is the probability that a
car can be assembled at this plant in a period of time
a) less than 19.5 hours? b) between 20 and 22 hours?
5 STATISTICAL INFERENCES
5.1 Introduction
In research, one always has some fixed ideas about certain population parameters based on
say, prior experiments, surveys or experience. However, these are only ideas. There is
therefore a need to ascertain whether these ideas /claims are correct or not.
The ascertaining of claims is done by first collecting information in the form of sample data.
We then decide whether our sample observations (statistic) have come from a postulated
population or not.
Definitions
A hypothesis is a claim (assumption) about a population parameters such as the population
mean, the population proportion or the population standard deviation is a postulated or a
stipulated value of a parameter
Example: The mean monthly cell phone bill in this city is μ= $42
The proportion of adults in this city with cell phones is π = 0.68
On the basis of observation data, one then performs a test to decide whether the postulated
hypothesis should be accepted or not. However, we note that the decision aspect is prone to
error/risk.
Null Hypothesis (denoted H0): Statement of zero or no change and is the hypothesis which is
to be actually tested for acceptance or rejection. If the original claim includes equality (<=, =,
or >=), it is the null hypothesis. If the original claim does not include equality (<, not equal,
>) then the null hypothesis is the complement of the original claim. The null hypothesis
always includes the equal sign. The decision is based on the null hypothesis.
Eg: The average number of TV sets in U.S. Homes is equal to three ( H 0 : μ  3 )
It’s always about a population parameter, and not about a sample statistic
Ie H 0 : μ  3 but NOT H 0 : x  3
We begin with the assumption that the null hypothesis is true
- Similar to the notion of innocent until proven guilty
Alternative Hypothesis (denoted H1 or Ha): Statement which is true if the null hypothesis is
false. it Challenges the status quo. It Is generally the hypothesis that the researcher is trying to
prove and it is accepted when H 0 is rejected and vice versa. The type of test (left, right, or
two-tail) is based on the alternative hypothesis.
5.2 The Hypothesis Testing Process

Claim: The population mean age is 50. Ie Hypothesis H 0 : μ  50 , vs H1: μ ≠ 50
Sample the population and find sample mea. Suppose the sample mean age was x  20 . This
is significantly lower than the claimed mean population age of 50. If the null hypothesis were
true, the probability of getting such a different sample mean would be very small, so you
reject the null hypothesis .In other words, getting a sample mean of 20 is so unlikely if the
population mean was 50, you conclude that the population mean must not be 50.
It is unlikely that you

would get a sample mean When infact this were the
of this value true population mean
- If the sample mean is close to the assumed population mean, the null hypothesis is not
rejected.
- If the sample mean is far from the assumed population mean, the null hypothesis is
rejected.
How far is “far enough” to reject H0? The critical value of a test statistic creates a “line in the
sand” for decision making -- it answers the question of how far is far enough.
Critical values
“Too Far Away” From Mean of Sampling Distribution
5.2.1 Possible Errors in Hypothesis Test Decision Making

When taking a decision about the acceptance or rejection of a null hypothesis/ alternative
hypothesis, there is a risk of committing an error. These errors are of two types:
Type I error; Mistake of rejecting the null hypothesis when it is true (saying false when
true). It is usually the more serious error.
The probability of a Type I Error is (denoted ) is Called the level of significance of the test
and it is Set by researcher in advance. = 0.05 and  = 0.01 are common. If no level of
significance is given, use  = 0.05. The level of significance is the complement of the level of
confidence in estimation.
Type II error: Mistake of failing to reject the null hypothesis when it is false (saying true
when false). The probability of a Type II Error is denoted by β
Remarks
1) The confidence coefficient (1-α) is the probability of not rejecting H0 when it is true.
2) The confidence level of a hypothesis test is 100 (1-α)%.
3) The power of a statistical test (1-β) is the probability of rejecting H0 when it is false
Possible Hypothesis Test Outcomes
Actual Situation
Decision H0 True H0 False
Do Not Reject H0 No Error Type II Error
Probability 1 - α Probability β
Reject H0 Type I Error No Error
Probability α Probability 1 - β
5.2.1 Relationship between Type I & Type II Error

Type I and Type II errors cannot happen at the same time
- A Type I error can only occur if H0 is true
- A Type II error can only occur if H0 is false
If Type I error probability (  ) increases, then Type II error probability ( β ) decreases
5.2.3 Level of Significance and the Rejection Region

Critical region: Set of all values which would cause us to reject H0
Critical value(s): The value(s) which separate the critical region from the non-critical region.
The critical values are determined independently of the sample statistics.
H0: μ = 3
H1: μ ≠ 3
This is a two-tail test because there is a rejection region in both tails
Test statistic: Sample statistic used to decide whether to reject or fail to reject the null
hypothesis
Probability Value (P-value): The probability of getting the results obtained if the null
hypothesis is true. If this probability is too small (smaller than the level of significance), then
we reject the null hypothesis. If the level of significance is the area beyond the critical values,
then the probability value is the area beyond the test statistic.
Decision: A statement based upon the null hypothesis. It is either "reject the null hypothesis"
or "fail to reject the null hypothesis". We will never accept the null hypothesis.
Conclusion: A statement which indicates the level of evidence (sufficient or insufficient), at
what level of significance, and whether the original claim is rejected (null) or supported
(alternative).
5.2.4 Steps in Hypothesis Testing

Any hypothesis testing is done under the assumption that the null hypothesis is true.
Here are the steps to performing hypothesis testing
a) Write the null and alternative hypothesis.
b) Use the alternative hypothesis to identify the type of test.
c) specify the level of significance,  and find the critical value using the tables
d) Compute the test statistic
e) Make a decision to reject or fail to reject the null hypothesis.
f) Write the conclusion
Remarks
The first thing to do when given a claim is to write the claim mathematically (if possible),
and decide whether the given claim is the null or alternative hypothesis. If the given claim
contains equality, or a statement of no change from the given or accepted condition, then it is
the null hypothesis, otherwise, if it represents change, it is the alternative hypothesis.
The type of test is determined by the Alternative Hypothesis ( H1 )
Left Tailed Test

H1: parameter < value
Notice the inequality points to the left
Decision Rule: Reject H0 if t.s. < c.v.
Right Tailed Test

H1: parameter > value Notice
the inequality points to the right
Decision Rule: Reject H0 if t.s. > c.v.
Two Tailed Test

H1: parameter not equal to a value
Notice the inequality points to both sides
Decision Rule: Reject H0 if t.s. < c.v. (left) or
t.s. > c.v. (right)
If the test statistic falls into the non rejection region, do not reject the null hypothesis H0. If
the test statistic falls into the rejection region, reject the null hypothesis. Express the
managerial conclusion in the context of the problem
Conclusions are sentence answers which include whether there is enough evidence or not
(based on the decision) and whether the original claim is supported or rejected. Conclusions
are based on the original claim, which may be the null or alternative hypotheses.
5.3 Approaches to Hypothesis Testing

There are three approaches to hypothesis testing namely Classical Approach, p vale approach
and the confidence interval approach
5.3.1 The Classical Approach

The Classical Approach to hypothesis testing is to compare a test statistic and a critical value.
It is best used for distributions which give areas and require you to look up the critical value
(like the Student's t distribution) rather than distributions which have you look up a test
statistic to find an area (like the normal distribution).
The Classical Approach also has three different decision rules, depending on whether it is a
left tail, right tail, or two tail test.
One problem with the Classical Approach is that if a different level of significance is desired,
a different critical value must be read from the table.
5.3.2 P-Value Approach

The P-Value Approach, short for Probability Value, approaches hypothesis testing from a
different manner. Instead of comparing z-scores or t-scores as in the classical approach,
you're comparing probabilities, or areas.
The level of significance (alpha) is the area in the critical region. That is, the area in the tails
to the right or left of the critical values.
The p-value is the area to the right or left of the test statistic. If it is a two tail test, then look
up the probability in one tail and double it.
If the test statistic is in the critical region, then the p-value will be less than the level of
significance. It does not matter whether it is a left tail, right tail, or two tail test. This rule
always holds.
Reject the null hypothesis if the p-value is less than the level of significance.
You will fail to reject the null hypothesis if the p-value is greater than or equal to the level of
significance.
The p-value approach is best suited for the normal distribution when doing calculations by
hand. However, many statistical packages will give the p-value but not the critical value. This
is because it is easier for a computer or calculator to find the probability than it is to find the
critical value.
Another benefit of the p-value is that the statistician immediately knows at what level the
testing becomes significant. That is, a p-value of 0.06 would be rejected at an 0.10 level of
significance, but it would fail to reject at an 0.05 level of significance. Warning: Do not
decide on the level of significance after calculating the test statistic and finding the p-value.
Here are a couple of statements to help you keep the level of significance the probability
value straight.
The Level of Significance is pre-determined before taking the sample. It does not depend on
the sample at all. It is the area in the critical region,that is the area beyond the critical values.
It is the probability at which we consider something unusual.
The Probability-Value can only be found after taking the sample. It depends on the sample. It
is the area beyond the test statistic. It is the probability of getting the results we obtained if
the null hypothesis is true.
5.3.3 Confidence Intervals as Hypothesis Tests

Using the confidence interval to perform a hypothesis test only works with a two-tailed test.
a ) If the hypothesized value of the parameter lies within the confidence interval with a 1-
alpha level of confidence, then the decision at an alpha level of significance is to fail to
reject the null hypothesis.
b) If the hypothesized value of the parameter lies outside the confidence interval with a 1-
alpha level of confidence, then the decision at an alpha level of significance is to reject
the null hypothesis.
However, it has a couple of problems.
 It only works with two-tail hypothesis tests.
 It requires that you compute the confidence interval first. This involves taking a z-
score or t-score and converting it into an x-score, which is more difficult than
standardizing an x-score.
5.4 Testing a Single Mean

The value for all population parameters in the test statistics come from the null hypothesis.
This is true not only for means, but all of the testing we're going to be doing.
The following hypotheses are to be tested: H 0 :    0 vs H1 :   0 0r H 0 :   0 0r
H 0 :   0 Where  0 is some hypothesised value.
The statistic and the critical values depends on whether σ, is known or unknown.
5.4.1 Population Standard Deviation Known

If the population standard deviation σ, is known, then the population mean has a normal
distribution, and you will be using the z-score formula for sample means. The test statistic is
x
the standard formula you've seen before. z 
/ n
The critical value is obtained from the normal table.
Example Test at 5% level the claim that the true mean # of TV sets in US homes is equal to
3. Suppose the sample results are n = 100, x  2.84 (σ = 0.8 is assumed known)
Solution
State the appropriate null and alternative hypotheses
- H0: μ = 3 H1: μ ≠ 3 (This is a two-tail test)
Determine the appropriate technique
- σ is assumed known so this is a Z test.
Determine the critical values
- For  = 0.05 the critical Z values are ±1.96
Xμ 2.84  3
Compute the test statistic ZSTAT so the test statistic is: ZSTAT  σ/ n
 0.8/ 100
 2.0
Exercise 5.1
1. A simple random sample of 10 people from a certain population has a mean age of 27.
Can we conclude that the mean age of the population is less than 30? The variance is
known to be 20. Let = .05.
2. Bon Air Elementary School has 300 students. The principal of the school thinks that the
average IQ of students at Bon Air is at least 110. To prove her point, she administers an
IQ test to 20 randomly selected students. Among the sampled students, the average IQ is
108. Assuming variance is known to be 100, should the principal accept or reject her
original hypothesis? at 5% level of significance
3. Central bank believes that if consumer confidence is too high, the economy risks over
heating. Low confidence is a warning that rcession might be on the way. In either case,
the bank may choose to intervene by altering interest rates. The ideal value for the bank's
chosen measure is 50. We may assume the measure is normally distributed with standard
deviation 10. The bank takes a survey of 25 people. Which returned a sample mean of 54
for the index. What would you advice the bank to do? Use = .05.
4. A manager will switch to a new technology if the production process exceeds 80 units per
hour. The manager asks the company statistician to test the null hypothesis: H0: μ = 80
against the alternative hypothesis: H1: μ >80 If there is strong evidence to reject the null
hypothesis then the new technology will be adopted. Past experience has shown that the
standard deviation is 8. A data set with n = 25 for the new technology has a sample mean
of 83 Does this justify adoption of the new technology?
5.4.2 Population Standard Deviation Unknown

If the population standard deviation σ, is unknown, then the population mean has a student's t
distribution, and you will be using the t-score formula for sample means. The test statistic is
very similar to that for the z-score, except that sigma has been replaced by s and z has been
x
replaced by t. ie t 
s/ n
The critical value is obtained from the t-table. The degrees of freedom is n-1.
Example 1 A fertilizer mixing machine is set to give 12 kg of nitrate for every 100kg bag of
fertilizer. Ten 100kg bags are examined. The percentages of nitrate are as follows: 11, 14, 13,
12, 13, 12, 13, 14, 11, 12. Is there reason to believe that the machine is defective at 5% level
of significance?
Solution
Hypothesis H0: μ = 12 H1: μ ≠ 12 (This is a two-tail test)
σ is unknown so this is a t test. Use the unbiased estimator ie s 

 ( x  x) 2
n 1
Critical Region based on  = 0.05 and 9 degrees freedom
t9, 0.025  2.262 ie reject H0: μ = 12 if tc  2.262
From calculator x  12.5 and s  1.0801
x 12.5  12
Test statistic tc    1.4639
s / n 1.0801 / 10
Decision since tc  1.2639  2.262 , we fail to reject H0 and conclude that the machine is not
defective.
Example 2 The following figures give the end of year profits of ten randomly selected
Chemists in Nairobi county.
On the basis of this data, test whether the average profit is greater than 30M KSH at 1% level
of significance
Solution
Hypothesis H0: μ = 30 H1: μ > 30 (This is a 1-tail test)
σ is unknown so this is a t test. Use the unbiased estimator ie s 

 ( x  x) 2
n 1
Critical Region based on  = 0.01 and 9 degrees freedom
t9, 0.01  2.82 ie reject H0: μ = 12 if tc  2.82
From calculator x  29.415 and s  3.6601
x   29.415  30
Test statistic tc    0.51
s / n 3.6601 / 10
Decision since tc  0.51  2.82 , we don’t reject H0 and conclude that the average profit is not
greater than 30M KSH.
Exercise 5.2
1. Identify the critical t value for each of the following tests:
a. A two-tailed test with =0.05 and 11 degrees of freedom
b. A one-tailed test with =0.01 and n=17
2. Consider a sample with n  20 x  8.0 and s  2 Do the following hypothesis tests.
a) . H0: μ =8.7 H1: μ > 8.7 at =0.01
b) H0: μ =8.7 H1: μ ≠ 8.7 at =0.05
3. It is widely believed that the average body temperature for healthy adults is 98.6 degrees
Fahrenheit. A study was conducted a few years go to examine this belief. The body
temperatures of n = 130 healthy adults were measured (half male and half female). The
average temperature from the sample was found to be x  98.249 with a standard
deviation s = 0.7332. Do these statistics contradict the belief that the average body
temperature is 98.6? test at 1% level of significance
4. A study is to be done to determine if the cognitive ability of children living near a lead
smelter is negatively impacted by increased exposure to lead. Suppose the average IQ for
children in the United States is 100. From a pilot study, the mean and standard deviation
were estimated to be x  89 and s = 14.4 respectively. Test at 5% level whether there is a
negative impact.
5. The average cost of a hotel room in New York is said to be $168 per night. To determine
if this is true, a random sample of 25 hotels is taken and resulted in x  $172.5
and s  $15.40 . Test the appropriate hypotheses at  = 0.05.
6. A sample of eleven plants gave the following shoot lengths
Shoot length (cm) 10.1 21.5 11.7 12.9 14.8 11.0 19.2 11.4 22.6 10.8 10.2
An earlier study reported that the mean shoot length is 15cm. Test whether the
experimental data confirms the old view at 5% level of significance.
7. A simple random sample of 14 people from a certain population gives body mass indices
as shown in Table 7.2.1. Can we conclude that the BMI is not 35? Let = .05.
subject 1 2 3 4 5 6 7 8 9 10 11 12 13 14
BM 23 25 21 37 39 21 23 24 32 57 23 26 31 45
8. A company selling licenses for a franchise operation claims that, in the first year, the
yield on an initial investment is 10%. How should a hypothesis test be stated? If there is
strong evidence that the mean return on the investment is below 10% this will give a
cautionary warning to a potential investor. Therefore, test the null hypothesis: H0 : μ = 10
against the alternative hypothesis: H1 : μ < 10 From a sample of n = 10 observations, the
sample statistics are: x  8.82 x and s = 2.40
9. We know the distance that an athlete can jump is normally distributed but we do not
know the standard deviation. We record 15 jumps: 7.48 7.34 7.97 5.88 7.48 7.67 7.49
7.48 8.51 5.79 7.13 6.80 6.19 6.95 5.93 Test whether these values are consistent with
a mean jump length of 7m. Do you have any reservations about this test?
10. The manufacturing process should give a weight of 20 ounces. Does the data show
evidence that the process is operating correctly? Test the null hypothesis: H0 : μ = 20 the
process is operating correctly against the alternative: H1 : μ ¹ 20 the process is not
operating correctly From the data set, the sample statistics are: n = 9, x  20.356
(ounces) and s = 0.6126

Sta 2200 Notes

Uploaded by

Copyright:

Available Formats

Sta 2200 Notes

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sta 2200 Notes

Uploaded by

Copyright:

Available Formats

Time is precious, but we do not know yet how precious it really is.

STA 2200 PROBABILITY AND STATISTICS II

Further Reference Text Books And Journals:

1.2 Discrete Random Variables and Probability Mass Function

4 5 6 7 8 This can also be written as a function

P 3  w  0  P(W  3)  P(W  2)  P(W  1)  0.65

P1  Y  3  P(Y  1)  P(Y  2)  352  355  15

Find the value of the constant c and PX  3

1.3 Continuous Random Variables and Probability Density Function

P 1  X  1  P 1  X  0  P0  X  1  0  14  14

Now P50  X  150  0.01 e 0.01x dx   e 0.01x

PX  100  0.01 e 0.01x dx   e 0.01x

Example 3 A continuous random variable X has a probability density function given by

P1  X  2.5  P1  X  2  P2  X  2.5   12 x  34 dx  14  x 4_ 3x 2

7) The pdf of a random variable X is given by f ( x)   1

1.4 Distribution Function of a Random Variables

Here t is introduced to facilitate summation /integration//

Properties of any cumulative distribution function

a) Show that k=13 b) Find the pmf of x

b) Find, in full, the probability density function of the random variable X.

1.5 Derived Random Variables

PY = 0  PX = 2  15 PY = 1  PX = 1  PX = 3  152  154  52

4. Let the pmf of a r.v X be given by f( x)   2 , determine the pmf of

5. Suppose the r.v X has a pmf given by f( x)   , Obtain the pmf

1.6 Change of Variable Technique

1.7 Expectation and Variance of a Random Variable

1.7.1 Expected Values

(i) E(c) = c, where c is an arbitrary constant.

(ii) Eax  b   (ax  b) P( X  x)dx   axP( X  x)dx   bP( X  x)dx

a  xP( X  x)dx  b  P( X  x)dx  a  b

(iii) Ekg(x)    kg( x) P( X  x)dx  k  g ( x) P( X  x)dx  kEg(x) 

1.7.2 Variance and Standard Deviation

5. A random variable X has the probability distribution shown below,

1.8 Mode, Median, Quartiles and Percentiles

b) At median m, F(m)  m2  0.5  m  0.5  1

c) At upper quartile u F(Q3 )  Q3  0.75  Q3  0.75 

2.1.1 Bernoulli distribution

2.1.2 Binomial Distribution

2.1.3 Poisson distribution

2.1.4 Geometric Distribution

Example 1 A sharpshooter normally hits the target 70% of the time.

a) P( X  2)  p1     0.70.3  0.21

2.1.5 The negative binomial distribution

2.1.6 Hyper geometric Distribution

 0.3398  0.3974  0.1975  0.9347  P(Y  2)  0.0653

2.2 Continuous Distribution

The mean and Variance of X are given by  

2.2.2 Exponential Distribution

Definition: A RV X is said to have an exponential distribution with parameter   0 if

4. Break downs occur on an old car with rate λ= 5 break-downs/month. The

2.2.3 Gamma Distribution

Definition A random variable X has gamma density if it pdf is given by

b) If   1 then we have exponential density function.

Example 1 Suppose the reaction time of a randomly selected individual to a certain

2.2.4 Beta Distribution

iii) B  ,    2 cos 2 1 t sin 2  1 tdt ie by putting x  cos 2 t in the function

3. MOMENTS AND MOMENT-GENERATING FUNCTIONS

tx (tx ) 2 (tx )3 (tx ) 4

all x all x 2! all x 3! all x

E (Y )  M ' ' (0) 