Lecture 4: Probability Distributions

Discrete Random variables and Discrete

Probability Distributions

Instructor :
 Discrete Probability Distributions
• Discrete Random variables and Discrete
Probability Distributions
• Mean, Variance, and Standard Deviation of
Discrete Random variables
• The Binomial Distribution
• Other Type of Distributions

2
 Random Variables
Random variable:
a numerical description of the outcome of an
experiment.

 Random variables are denoted by capital letters, X,

Y, …, specific values by lower case letters, x, y, …

 Discrete random variable:

The number of possible outcomes can be counted.
 Continuous random variable:
Outcomes over one or more continuous intervals of real
numbers.

3
 Random Variables
• A random Variable is a variable whose values are
determined by chance, and it is a function from
the sample space of a probability experiment to
the real numbers.
X :S  R
Such that {s : X (s ) x } is an event x  R , and s  S
• That is, a R.V. is a function that assumes values with certain
probabilities.
• If a sample space S is discrete, then every R.V. defined on S
is also discrete, i.e. its range is countable.
• A random variable is either discrete or continuous; in this
lecture, we focus on the discrete case.

4
 Examples of Random Variables
• Experiment: flip a coin 3 times.
– Outcomes: TTT, TTH, THT, THH, HTT, HTH, HHT, HHH
– Random variable: X = number of heads.
– X can be either 0, 1, 2, or 3.
• Experiment: observe end-of-week closing stock price.
– Random variable: Y = closing stock price.
– X can be any nonnegative real number.
• The outcome of rolling a single die.
• The number of boys in a family with three children.
• The number of heads that appear when a coin is flipped ten
times.
• The number of A students in a randomly-selected sample of
10 students.
5
 Discrete Probability Distributions
• A discrete probability distribution is a listing of all possible
values of a random variable X along with their probabilities
P(X). The probabilities are determined theoretically or by
observation. X | x1 x2 ... xk (...)
,
P ( X ) | p1 p2 ... pk (...)
p
k 1
k 1.

• Any probability distribution should satisfy 2 conditions

 P(x) = 1.
i.e. , The sum of the probabilities of all the events in the sample space must equal 1

2. 0 ≤ P(x) ≤ 1
i. e,. The probability of each event in the sample space must be between or equal
to 0 and 1.

6
 Example 1
Determine whether each distribution is a probability distribution

X 0 5 10 15 20 X 1 2 3 4
P(X) 1/5 1/5 1/5 1/5 1/5 P(X) 1/4 1/8 1/16 9/16

Yes, it is a probability Yes, it is a probability

distribution distribution
X 0 2 4 6 X 2 3 7
P(X) -1 1.5 0.3 0.2 P(X) 0.5 0.3 0.4

No, it is not a probability distribution, No, it is not a probability distribution,

since P(X) cannot be 1.5 0r -1. since  P(x) = 1.2.

7
 Example 2
a. Construct a probability distribution for rolling a single die.
Solution
Since the sample space is 1, 2, 3, 4, 5, 6 and each outcome has a
probability of 1/6. Then distribution of the outcome of rolling a
single die is as shown.
Outcome X 1 2 3 4 5 6
Probability P(X) 1/6 1/6 1/6 1/6 1/6 1/6

b. Construct a probability distribution for the number of girls in a

family with three children
Solution
X 0 1 2 3
P(X) 1/8 3/8 3/8 1/8

8
 Example 3
Represent graphically the probability distribution for the sample
space for tossing three coins
Solution
When three coins are tossed, the sample space is represented as
S = { TTT, TTH, THT, HTT, HHT, HTH. THH, HHH }
If X is the random variable for the number of heads.
Then X assumes thee values 0, 1, 2, and 3.
Probabilities for the values of X can be determined as follows:

No Heads One Head Two Heads Three Heads

TTT TTH THT HTT HHT HTH THH HHH

1/8 1/8 1/8 1/8 1/8 1/8 1/8 1/8

1/8 3/8 3/8 1/8

9
Hence, the probability of getting:
no heads is 1/8, one heads is 3/8, two heads is 3/8, and three
heads is 1/8. From these values, a probability distribution can
be constructed by listing the outcomes and assigning the
probability of each outcome, as shown here.
Number of head X 0 1 2 3
Probability P(X) 1/8 3/8 3/8 1/8

The Graph
Probability P(X)

3/8 -

2/8 -
1/8 -

X
0 1 2 3

10
 Example 4
Represent graphically the probability distribution of rolls of two
dice

11
 Expected Value (Average) of a Discrete
Random Variable
• Each time a random experiment is carried out, an
associated R.V. takes on a value. Is it possible find an
“average” value of the R.V. ?
• For instance, on average, how many bits are received in
error when 8 bits are transmitted?
• Or, on average, how many bits must be received until the
first (the r-th) error occurs?
– Note that in both cases, you can take a sample, i.e. repeat the respective
experiment several times and take the value of the RV each time.
– An average computed this way is, however, dependent on the sample!
– Is it possible to get a sample independent value?
• Expected value of a random variable X is the theoretical
analogy of the mean, or weighted average of possible values:

E[X] =  x i f(x i )
i =1
12
 Mean or Expectation (Expected Value)
The Mean or expected value (expectation) of a discrete
random variable X is given by

 E(X )  x k p k .
k 1

The formula gives a weighted average of values of a

random variable, with the probabilities being the weights of
the respective values.

13
 Properties of Expectation
1. If a and b are constants, then
E(aX  b ) a E(X )  b .
and
E(aX ) a E(X ).
E(b ) b .
2. If X and Y are random variables, then
E(X Y ) E(X )  E(Y ).
3. If g(X) and h(X) are two functions of random
variables X, then
E( g X   h X ) E( g X )  E(h X ).

14
 Example 5
Find the mean of the number of spots that appear when a die is
tossed.

Solution

That is, when a die is tossed many times, the theoretical

mean will be 3.5.

15
 Example 6
The probability distribution shown represents the number of trips
of five nights or more that American adults take per year. Find
the mean.

Solution

Hence, the mean of the number of trips lasting five nights

or more per year taken by American adults is 1.2.

16
 Variance and Standard Deviation
The average spread of a D.R.V. around its expectation is called
the standard deviation, and is computed via the so-called
variance: 
 V ar (X ) E (x k   )  (x k   ) p k E x k    E x 
2 2 2 2 2

k 1

  x k2 p k   2
2
Variance
k 1

   2
 k k
x 2

k 1
p   2
Standard deviation

17
 Properties of Variance

1. If a and b are constants, then

Var(aX  b ) a 2 Var(X ).
and
Var(aX ) a 2 Var( X ).
Var(b ) 0.
2. If X and Y are random variables, then
Var(X  Y ) Var(X )  Var(Y ).
3. Let X be a random variable with probability
distribution f(x), the variance of the function g(X)
is 2
Var( g X ) E( g X  )   E( g X )  .
2

18
 Example 7
Compute the mean, variance and standard deviation of the
number of spots that appear when a die is tossed.
Solution
The probability distribution of the toss of a die, shown here.
Outcome X 1 2 3 4 5 6
Probability P(X) 1/6 1/6 1/6 1/6 1/6 1/6

The mean   X . P X 
1 1 1 1 1 1 21
1.  2.  3.  4.  5.  6.  3.5.
6 6 6 6 6 6 5
The Variance
 E  X    E  X    X
2
2 2 2
. P  X    2
 1 1 1 1 1 1 21
 12.  22.  32.  42.  52.  62.   3.5   2.9.
2

 6 6 6 6 6 6 5
The standard deviation
   2  2.9 1.7

19
 Example 8 Selecting Numbered Balls
A box contains 5 balls. Two are numbered 3, one is numbered 4,
and two are numbered 5. The balls are mixed and one is selected
at random. After a ball is selected, its number is recorded. Then
it is replaced. If the experiment is repeated many times, find the
variance and standard deviation of the numbers on the ball.
Solution
Let x be the number on each ball. The probability distribution is.
Number on ball X 3 4 5
Probability P(X) 2/5 1/5 2/5
The mean
2 1 2
  X . P  X  3.  4.  5. 4.
5 5 5
The Variance
 2   X 2 . P  X    2
 2 1 2 4
 32.  42.  52.   4   0.8.
2

 5 5 5 5
The standard deviation
   2  0.8 0.894.
20
 Example 9
A talk radio station has four telephone lines. If the host ‫ المضيف‬is
unable to talk (i.e., during a commercial) or is talking to a person,
the other callers are placed on hold. When all lines are in use,
others who are trying to call in get a busy signal. The probability
that 0, 1, 2, 3, or 4 people will get through is shown in the
distribution. Find the variance and standard deviation for the
distribution.

Should the station have considered getting more phone lines

installed?

21
 Solution

No need to install new phone lines. The mean number of

people calling at any one time is 1.6. Since the standard
deviation is 1.1, At least 75% of the callers would be
accommodated by having four phone lines because;
μ + 2σ = 1.6 + 2(1.1) = 3.8
(Review Chebyshev’s theorem)
22
 Example 10
You play a lottery in which you buy a ticket for $50 and are told
you have a 1 in 1000 chance of winning $25,000. The random
variable X is your net winnings, and its probability distribution is
x f(x)
-$50 0.999
$24,950 0.001
Solution
•E[X] = -$50(0.999) + $24,950(0.001) = -$25.00
•Var[X] = (-50 - [-25.00])2(0.999) + (24,950 - [-
25.00])2(0.001) = 624,375

23
 Example 11 Winning Tickets
One thousand tickets are sold at $1 each for four prizes of $100,
$50, $25, and $10. After each prize drawing, the winning ticket is
then returned to the pool of tickets. What is the expected value if
you purchase two tickets?
Solution
The problem can be set up as follows:
Gain X $98 $48 $23 $8 -$2
Probability P(X) 2/1000 2/1000 2/1000 2/1000 992/1000

The Expected value

E  X    X . P  X 
2 2 2 2 992
$98.  $48.  $23  $8  (  $2).  $1.63.
1000 1000 1000 1000 1000
An Alternate solution is
2 2 2 2
E  X  $100.  $50.  $25  $10  $2  $1.63.
1000 1000 1000 1000

24
 Example 12 Bond Investment
A financial adviser suggests that his client ‫عميل‬select one of two
types of bonds ‫سندات‬in which to invest $5000. Bond X pays a
return of 4% and has a default rate of 2%. Bond Y has 2.5%
return a default rate of 1%. Find the expected rate of return and
decide which bond would be a better investment. When the bond
defaults, the investor loses all the investment.
Solution
- The return on bond X is: $5000. 4% = $200.
The Expected value is
E  X  $200 0.98   $5000 0.02  $96.
- The return on bond Y is: $5000. 2.5% = $125.

The Expected value is

E  X  $125 0.99   $5000 0.001 $73.75.
Hence, bond X would be a better investment since the expected return is
higher.
25
 Example 13
One thousand tickets are sold at $1 each for a color television
valued at $350. What is the expected value of the gain if you
purchase one ticket?

Solution
The problem can be set up as follows:

26
 Example 14
A construction company wants to submit a bid ‫ تقديم عطاء‬for
remodeling a school. The research and planning needed to make
the bid cost $4000. If the bid were accepted, the company would
make $26,000. Would you advise the company to spend the
$4000 if the bid has only 20% probability of being accepted?
Explain your reasoning.

Solution

27
 Example 15
A dairy farmer ‫مزرعة مخصصة لمنتجات األلبان‬estimates for the
next year the farm’s cows ‫ بقرة‬will produce about 25,000
gallons of milk. Because of variation in the market price of milk
and cost of feeding the cows, the profit per gallon may vary
with the probabilities given in the table below. Estimate the
profit on the 25,000 gallons.

Solution

28
 Exam Question
Table 1 gives the number of job applications processed at a small employment
agency during the past 100-day period.
Table 1: Number of Job Applications Processed during the Past 100-Day Period

Number of Job Applicants Number of Days Achieved

7 10
8 10
10 20
11 30
12 20
14 10
100

Determine the expected number of applications processed and the variance

and standard deviation.
Solution
To the extent that we believe that the experience of the past 100 days is
typical, we can find the relative frequency distribution and equate its
probability distribution. This and the other calculations to find E(X) and Var X
are shown in Table 2.
29
Table 2: Calculations to Find the Expected Value and Variance

30
 Notes on Discrete R.V.
• We notice that
– The discrete distributions are described by their
Cumulative Probability Fun CPFs;
– The CPFs are non-decreasing step functions;
– The CPFs only make jumps at points in the range of the
random variable.
– One can recover the respective mass functions using the
size of the jump;
– The expectation and the variance are computed using the
respective mass function;
– The calculation of probabilities, expectation and variance
involves summation.

31
Any Question
Thank You