IT2011

Key Concepts

• Element is the source of relevant information or data such as individual, entity, or population unit.
• A variable is the characteristic or property of the elements.
• A random variable is a variable being measured to produce numerical observations associated
with the random outcomes of a chance experiment. Random variables are typically named with
capital letters (X, Y, Z, etc.)
• Discrete Random Variables are random variables where the observed numerical values are
produced by counting and assumes whole numbers only.
• Continuous Random variables are random variables where observed numerical values are
results of measuring and may take on any numbers contained within any numerical intervals.

Discrete Random Variables and Its Probability Distribution

Properties of Probability Distribution for Discrete Random Variables

1. In discrete probability distribution, the probability values for all its possible outcomes are greater
than or equal to zero [𝑃(𝑥) ≥ 0].
2. The sum of the probability values associated with the corresponding outcome is equal to 1
[∑𝑃(𝑥) = 1].
3. The probability mass function is an equation that defines the relationship between an outcome and
[(𝑥):𝑛]
its associated probability [𝑃(𝑥) = ].
𝑁
4. The probability value is the quotient of the frequency of occurrences divided by the total number of
occurrences.
5. From the given example below, since the probability is a measure of chance, the probability value
of 0.60 for (x)= 2 [two (2) heads] would mean that there is a 60% chance to obtain 2 heads in
tossing a coin three (3) times.

Example 1

Assume that a coin is was tossed three times. Let the random variable X be the number of heads. The following
table summarizes the results and shows the probability distribution of discrete random variable.

Outcomes (x) Number of Occurrences Probability Mass function Probability Value

[(x):n] [(𝑥): 𝑛] P(x)
𝑁
0 1 {TTT} 1 0.125
8
1 3 {HTT, THT, TTH} 3 0.375
8
2 3 {THH, HTH, HHT} 3 0.375
8
3 1 {HHH} 1 0.125
8
Table 1. Probability Distribution of Discrete Random Variable X

Interpretation:

The table above shows that in tossing a coin thrice, there are 8 possible outcomes. An outcome with three (3)
heads or zero (0) head can both occur in one (1) way, with probability values of 0.125. An outcome with two (2)
heads or one (1) head can both occur in three (3) ways so both have probabilities of 0.375.

03 Handout 1 *Property of STI

 [email protected] Page 1 of 7
 IT2011

Steps to Represent a Distribution Graphically Using a graph

1. Assign the probability values on the vertical grid.

2. Assign the discrete random variable (x) on the horizontal grid.
3. Draw a rectangular bar on each section on the horizontal grid with length corresponds to its
probability value.

Use the given distribution in Example 1

0.4 0.375 0.375

0.35
0.3
Probability Value

0.25
0.2
0.15 0.125 0.125
0.1
0.05
0
0 1 2 3
Number of Heads

Figure 1. Graphical Presentation of the Probability Value of the Number of Heads in Tossing a Coin Three
(3) times

Mean, Variance, and Coefficient of variation of Discrete Random Variables

Population Mean or Expected Value

To compare two or more elements, determine the “expected value” of the random variable denoted
by E(x) by calculating the population mean (𝜇). A parameter or a population characteristic that describes
the center or common data in the distribution.

To compute the population mean/expected value,

1. Multiply each of the values of x by the associated probability. (𝑥) ∙ 𝑃(𝑥).

2. The population mean is the sum of the products. 𝜇 = ∑[(𝑥) ∙ 𝑃(𝑥)] = 𝐸(𝑥)

Variance

Measuring variability will help in comparing the consistency of the average. Population variance is
a parameter that measures the average squared distance or deviation of each item in the data from the
mean.

To measure the variability of the random variable, compute the population variance (𝜎 2 ).

1. Subtract the population mean (𝜇) from each of the values (x) to get (𝑥 − 𝜇).
2. Square their differences (𝑥 − 𝜇)2 .
3. Multiply each squared difference to its associated probability (𝑥 − 𝜇)2 ∙ 𝑃(𝑥).
4. The population variance (𝜎 2 ) is equal to the sum product of the squared difference and its
probability. 𝜎 2 = ∑[(𝑥 − 𝜇)2 ∙ 𝑃(𝑥)]

03 Handout 1 *Property of STI

 [email protected] Page 2 of 7
 IT2011

Note: The lesser the variability, the more consistent is its average.

Coefficient of Variation

It is best to determine the coefficient of variance (CV) for more substantial interpretation and
decision-making. The coefficient of variation is the percentage variability of the mean.

To determine the coefficient of variation,

1. Calculate the standard deviation (𝜎) by finding the square root of variance. (𝜎 = √𝜎 2 )
2. Divide the standard deviation (𝜎) by the population mean (𝜇) then multiply by 100%.
𝜎
𝐶𝑉 = ∙ 100%
𝜇

Example 2

Suppose a sales manager have to assess his agents for the “Best Agent of the Year”. The number of products sold
by the agents will be the basis of selection. The table below shows the number of products sold by two (2) agents
for 40 consecutive days.

Agent A Agent B
Number of products sold Frequency P(x) Frequency P(x)
(x)
0 1 0.03 1 0.03
1 2 0.07 4 0.13
2 3 0.10 12 0.41
3 6 0.20 7 0.23
4 10 0.33 4 0.13
5 8 0.27 2 0.07
Total: N = 30 N = 30
Table 1. Probability Distribution of the Number of Products Sold by Two Agents

Consider the table below to find the expected value and variance of each set of distribution.
No. of Product Probability
(𝑥) ∙ 𝑃(𝑥) (𝑥 − 𝜇) (𝑥 − 𝜇)2 (𝑥 − 𝜇)2 ∙ 𝑃(𝑥)
sold (x) P(x)
0 0.03 0 -3.54 12.53 0.38
1 0.07 0.07 -2.54 6.45 0.45
2 0.10 0.2 -1.54 2.37 0.24
Agent A
3 0.20 0.6 -0.54 0.29 0.06
4 0.33 1.32 0.46 0.21 0.07
5 0.27 1.35 1.46 2.13 0.58
𝜇 = ∑[(𝑥) ∙ 𝑃(𝑥)] 𝜎 2 = ∑[(𝑥 − 𝜇)2 ∙ 𝑃(𝑥)]
=3.54 =1.77
0 0.03 0 -2.51 6.30 0.19
1 0.13 0.13 -1.51 2.28 0.30
2 0.41 0.82 -0.51 0.26 0.11
Agent B
3 0.23 0.69 0.49 0.24 0.06
4 0.13 0.52 1.49 2.22 0.29
5 0.07 0.35 2.49 6.20 0.43
𝜇 = ∑[(𝑥) ∙ 𝑃(𝑥)] 𝜎 2 = ∑[(𝑥 − 𝜇)2 ∙ 𝑃(𝑥)]
=2.51 = 1.37

03 Handout 1 *Property of STI

 [email protected] Page 3 of 7
 IT2011

Interpretation:

• The average number of products sold by Agent A and agent B is 3.54 and 2.51 respectively.
• Agent B has lesser variability than Agent A.
• Since Agent B has lesser variability than Agent A, therefore the average of Agent B is more consistent than
the average of Agent A.
• The standard deviation of the number of products sold by Agent A and Agent B is √1.77 = 1.33 and
√1.37 = 1.17, respectively.
1.33
• The coefficient of variation of the number of products sold by Agent A and Agent B is = 0.38 = 38%
3.54
1.17
and = 0.47 = 47%, respectively.
2.51
• Using CV as a measure of variability, it reveals the size of consistency relative to the size of the mean (in
percentage), therefore, Agent A is the Best Agent of the year.

Continuous Random Variable and Its Probability Density

Function
A continuous random variable is a variable whose range of possible values is an infinite collection
containing as many elements as there are real numbers and has a probability density function f(x) that can
be used to compute 𝑃(𝑎 < 𝑋 < 𝑏) by getting the area bounded above by f, bounded below by the 𝑥 − 𝑎𝑥𝑖𝑠,
bounded on the left by the line 𝑥 = 𝑎 and bounded on the right by the line 𝑥 = 𝑏. For a continuous variable,
the probability that 𝑋 takes on any particular value 𝑥 is 0, that is 𝑃(𝑋 = 𝑥) = 0.

Figure 2. The shaded region represents 𝑃(𝑎 < 𝑋 < 𝑏)

Source: https://intellipaat.com/wp-content/uploads/2015/09/probability-for-a-continuous-random-variable.png

Probability Density Function

The probability density function (pdf) of a continuous random variable 𝑋 is an integrable function 𝑓(𝑥)
satisfying the following:

1. 𝑓(𝑥) is a positive everywhere bounded by 𝑎 < 𝑥 < 𝑏, that is, 𝑓(𝑥) > 0, for all 𝑎 < 𝑥 < 𝑏.

03 Handout 1 *Property of STI

 [email protected] Page 4 of 7
 IT2011

2. The area under the curve 𝑓(𝑥) bounded by 𝑎 < 𝑥 < 𝑏, is one (1), that is:
𝑏
∫ 𝑓(𝑥)𝑑𝑥 = 1
𝑎

3. If 𝑓(𝑥) is the probability density function of x, then the probability of 𝑎 < 𝑥 < 𝑏 is given by the
integral of 𝑓(𝑥) from 𝑎 to 𝑏, that is:
𝑏
𝑃(𝑎 < 𝑥 < 𝑏) = ∫ 𝑓(𝑥)𝑑𝑥
𝑎

Example 3:

Let 𝑋 be a continuous random variable whose probability density function is

3𝑥 2 , 0<𝑥<1
𝑓(𝑥) = {
0, 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒
a. Verify if it is a valid probability density function; and
b. find the probability that X falls between zero (0) to 0.5? That is, what is 𝑃(0 < 𝑋 < 0.5)?

* Take note that, 𝑓(𝑥) ≠ 𝑃(𝑋 = 𝑥). For example, 𝑓(0.7) = 3(0.7)2 = 1.47, which is clearly not a probability.
In continuous random variable, 𝑓(𝑥) is not the probability but instead the height of the curve at 𝑋 = 𝑥. The
total area under the curve is always equal to one (1), and these areas defined the probabilities.

Solution

The graph of the given function is as shown below,

𝑃(0 < 𝑋 < 1)

The shaded part shows the area under the curve

or the probability that X falls between zero (0)
and one (1).

Figure 3. Probability Density Function

a. To verify if the given function is a valid probability function, it should satisfy the properties of pdf.

The given function is:

3𝑥 2 , 0<𝑥<1
𝑓(𝑥) = {
0, 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒

03 Handout 1 *Property of STI

 [email protected] Page 5 of 7
 IT2011

Continuation of Example 3

1. 𝑓(𝑥) > 0
1
2. 𝑃(0 < 𝑥 < 1) = ∫0 3𝑥 2 𝑑𝑥 = [𝑥 3 ] 10 = 1 − 0 = 1

Since the function 𝑓(𝑥) > 0, for all 0 < 𝑥 < 1, and the area under the curve 𝑓(𝑥) bounded by
0 < 𝑥 < 1, is equal to one (1), then it is a valid probability density function.

b. To find the probability that X falls between zero (0) and 0.5, 𝑃(0 < 𝑥 < 0.5), integrate 𝑓(𝑥) from 0 to
0.5, that is:

0.5
𝑃(0 < 𝑥 < 0.5) = ∫ 3𝑥 2 𝑑𝑥
0
0.5
= [𝑥 3 ]
0
= 0.125 − 0
= 0.125

Therefore, 𝑃(0 < 𝑥 < 0.5) = 0.125

𝑃(0 < 𝑋 < 0.5)

Figure 4. Shaded region represents 𝑃(0 < 𝑋 < 0.5)

𝑃(0 < 𝑋 < 0.5)

Expected Value and Variance of Continuous Random Variable

The expected value or mean of a continuous random variable is the probability density-weighted integral
of all possible values.

The expected value of a continuous random variable X is:

∞

∫ 𝑥 ∙ 𝑓(𝑥)𝑑𝑥
−∞

provided that the integral exists.

The variance of a continuous random variable 𝑋 is:

∞
2 2
𝜎 = 𝑉𝑎𝑟(𝑥) = 𝐸[(𝑋 − 𝜇) ] = ∫(𝑥 − 𝜇) ∙ 𝑓(𝑥)𝑑𝑥
−∞

and this can be simplified as:

∞
𝟐)
𝑬(𝑿 = ∫ 𝒙𝟐 ∙ 𝒇(𝒙)𝒅𝒙
−∞

03 Handout 1 *Property of STI

 [email protected] Page 6 of 7
 IT2011

Example 4

Consider the random variable 𝑋 in example number three (3) with probability density function

3𝑥 2 , 0<𝑥<1
𝑓(𝑥) = {
0, 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒
The expected value or mean of the continuous Random variable is:
∞

𝜇 = 𝐸(𝑥) = ∫ 𝑥 ∙ 𝑓(𝑥)𝑑𝑥
−∞

= ∫ 𝑥 ∙ 3𝑥 2 𝑑𝑥
0

= ∫ 3𝑥 3 𝑑𝑥
0

3𝑥 4 1
=[ ]
4 0

𝟑
=
𝟒
The variance is
∞
2)
𝐸(𝑋 = ∫ 𝑥 2 ∙ 𝑓(𝑥)𝑑𝑥
−∞

= ∫ 𝑥 2 ∙ 3𝑥 2 𝑑𝑥
0

= ∫ 3𝑥 4 𝑑𝑥
0

3𝑥 5 1
=[ ]
5 0

𝟑
=
𝟓

References:

PennState (2020). Continuous distribution. Retrieved from

https://online.stat.psu.edu/stat414/lesson/14/14.1

03 Handout 1 *Property of STI

 [email protected] Page 7 of 7