Applied Statistics -Notes-2 Dr.

T K Datta

Total Probability Rule:

Let 𝐸1 , 𝐸2 , … . . 𝐸𝑛 be mutually exclusive and exhaustive events and B be another event. Then,
𝑃(𝐵) = 𝑃(𝐸1 ∩ 𝐵) + 𝑃(𝐸2 ∩ 𝐵) + ⋯ . 𝑃(𝐸𝑛 ∩ 𝐵)
= 𝑃(𝐸1 )𝑃(𝐵/𝐸1 ) + 𝑃(𝐸1 )𝑃(𝐵/𝐸2 ) + ⋯ . +𝑃(𝐸𝑛 )𝑃(𝐵/𝐸𝑛 ).
Example 1 (2-107): The probability is 2% that an electrical connector that is kept dry fails
during the warranty period of a portable computer. If the connector is wet, then the probability
of failure during warranty period is 6%. If 90% of the connector are kept dry and the remaining
10% are wet, what is the probability that a randomly selected connector will fail during the
warranty period?
Solution: Define: E1 – connector is kept dry; E2 – the connector is wet; F – the connector fails.
To find P(F).
Given: P(E1) = 0.9, P(E2) = 0.1, P(F/E1) = 0.02 and P(F/E2) = 0.06.
By total probability rule,
P(F) = P(E1) P(F/E1) + P(E2) P(F/E2) = 0.9*0.02 + 0.1*0.06 = 0.024.
Example 2: There are three identical boxes - Box1, Box 2 and Box 3. Each box contains red,
white and black balls. The details are given below:
BOX

1 2 3
Red 8 10 7
Ball White 4 6 5
Black 8 6 6
A box is selected at random and a ball is drawn from it. Compute the probability that the ball
drawn is white.
Solution: E1: Box 1 is selected, E2: Box 2 is selected, E3: Box 3 is selected and W: a white ball
is drawn.
P(W) = P(E1) P(W/E1) + P(E2) P(W/E2) + P(E3) P(W/E3)
= 1/3* 4/20 + 1/3*6/22 + 1/3*5/18 = 0.2502.

Bayes’ Theorem: Let 𝐸1 , 𝐸2 , … . . 𝐸𝑛 be mutually exclusive and exhaustive events and B be

another event. Then,
𝑃(𝐸𝑘 )𝑃(𝐵/𝐸𝑘 ) 𝑃(𝐸𝑘 )𝑃(𝐵/𝐸𝑘 )
P(Ek/B) = = = 𝑃(𝐸 , k = 1, 2, …, n.
𝑃(𝐵) 1 )𝑃(𝐵/𝐸1 )+𝑃(𝐸2 )𝑃(𝐵/𝐸2 )+⋯+𝑃(𝐸𝑛 )𝑃(𝐵/𝐸𝑛 )

Example 1: The probability is 2% that an electrical connector that is kept dry fails during the
warranty period of a portable computer. If the connector is wet, then the probability of failure

1
 Applied Statistics -Notes-2 Dr. T K Datta

during warranty period is 6%. The 90% of the connector are kept dry and the remaining 10%
are wet. A randomly selected connector fails during the warranty period.
a) What is the probability that it was dry?
b) What is the probability that it was wet?
Solution: : Define: E1 – connector is kept dry; E2 – the connector is wet; F – the connector
fails.
To find P(F).
Given: P(E1) = 0.9, P(E2) = 0.1, P(F/E1) = 0.02 and P(F/E2) = 0.06.
P(F) = P(E1) P(F/E1) + P(E2) P(F/E2) = 0.9*0.02 + 0.1*0.06 = 0.024.
𝑃(𝐸1 )𝑃(𝐹/𝐸1 ) 𝑃(𝐸1 )𝑃(𝐹/𝐸1 )
a) By Bayes’ theorem, P(E1/F) = = 𝑃(𝐸 )𝑃(𝐹/𝐸
𝑃(𝐹) 1 1 )+𝑃(𝐸2 )𝑃(𝐹/𝐸2 )

0.018
= 0.024 = 0.75.
𝑃(𝐸2 )𝑃(𝐹/𝐸2 ) 𝑃(𝐸2 )𝑃(𝐹/𝐸2 )
b) By Bayes’ theorem, P(E2/F) = = 𝑃(𝐸
𝑃(𝐹) 1 )𝑃(𝐹/𝐸1 )+𝑃(𝐸2 )𝑃(𝐹/𝐸2 )

0.006
= 0.024 = 0.25.

Example 2: There are three identical boxes - Box1, Box 2 and Box 3. Each box contains red,
white and black balls. The details are given below:
BOX

1 2 3
Red 8 10 7
Ball White 4 6 5
Black 8 6 6
A box is selected at random and a ball is drawn from it. If the ball drawn is black, compute the
probability that the ball belongs to Box 3.
Solution: E1: Box 1 is selected, E2: Box 2 is selected, E3: Box 3 is selected and B: a black ball
is drawn.
P(B) = P(E1) P(B/E1) + P(E2) P(B/E2) + P(E3) P(B/E3)
= 1/3* 8/20 + 1/3*6/22 + 1/3*6/18 = 0.33535.
By Bayes’ theorem,
𝑃(𝐸3 )𝑃(𝐵/𝐸3 ) 𝑃(𝐸3 )𝑃(𝐵/𝐸3 )
P(E3/B) = 𝑃(𝐸 )𝑃(𝐵/𝐸 )+𝑃(𝐸
=
1 1 2 )𝑃(𝐵/𝐸2 )+𝑃(𝐸3 )𝑃(𝐵/𝐸3 ) 𝑃(𝐵)

= (1/3*6/18)/0.33535 = 0.3313.

2
 Applied Statistics -Notes-2 Dr. T K Datta

Random variables, probability mass functions, probability density

functions:
Definition: A random variable is a function that assigns a real value to any outcome in the
sample space of a random experiment. A random variable is denoted by uppercase letters, like
X, Y, ….. and the values are denoted by the corresponding lowercase letters x, y, ……
Examples:
• Number of heads in 10 tosses of a coin.
• Number of even faces in rolling a die 5 times.
• Rainfall in a particular state during monsoon.
Types of random variables: Random variables are either discrete or continuous.
Discrete random variable: A random variable is called a discrete random variable if it can
take a value either from a finite set of values or from a countably infinite set of value.
Continuous random variable: A random variable is called a continuous random variable if
it can take any real value from a given interval, finite or infinite.
Examples: Describe whether the following random variables are discrete or continuous:
a) The time until a projectile returns to the earth. (C)
b) The number of times a transistor in a computer memory changes its state in one
operation. (D)
c) The outside diameter of a machined shaft. (C)
d) The concentration of output from a reactor. (C)
e) The number of cracks in a 1.5 km stretch of a highway. (D)

Discrete Probability Distribution

Some examples of discrete random variable:

1. A coin is tossed 8 times. X denotes the number of tails in 8 tosses. Here, X is a discrete
random variable and I can take on values 0, 1, 2, ….8.
2. Suppose 5% of the printers of a supplier are defective. Let X denotes the number of
defectives out of 10 printers sold.
Probability mass function (pmf): The probability mass function of a discrete random variable
X is denoted by f(x) or p(x) and is defined by
f(x) = P(X =x) where x is any possible value of X.

3
 Applied Statistics -Notes-2 Dr. T K Datta

Properties of f(x): Let the discrete random variable X takes on values x1, x2,…,xn, then
1. f(xk) ≥ 0, k = 1, 2, …., n.
2. ∑𝑛𝑖=1 𝑓(𝑥𝑖 ) = 1.
3. 𝑃(𝑥𝑘 ) = 𝑃(𝑋 = 𝑥𝑘 ), k = 1, 2, …., n.
Discrete Probability Distribution: All values of a discrete random variable X with
corresponding probabilities constitutes the probability distribution of that discrete random
variable X. This is the list consisting of all possible X values with corresponding probabilities.
It can be represented by a table consisting of two rows showing the X values and corresponding
probabilities. Sometimes we can represent it by a function of x.
Example 1: Let X is a discrete random variable which can take on values 0, 1, 2, 3 with
corresponding probabilities 0.4, 0.1, 0.2 and 0.3, respectively. We can represent it in tabular
form as:
x 0 1 2 3
f(x) 0.4 0.1 0.2 0.3
This table defines the probability distribution of the discrete random variable x.
𝑥2
Example 2: f(x) = , x = 1, 3, 5.
35

Problem 1(3-15): Verify that the following function f(x) is a probability mass function. Also,
find the requested probabilities.
x -2 -1 0 1 2
f(x) 1/8 2/8 2/8 2/8 1/8

a) 𝑃(𝑋 ≤ 2); b) 𝑃(𝑋 > −2); c) 𝑃(−1 ≤ 𝑋 ≤ 1); d) 𝑃(𝑋 ≤ −1 𝑜𝑟 𝑋 = 2); e) 𝑃(𝑋 <
1 𝑎𝑛𝑑 𝑋 > −1)
Problem 2: Find the value of m for which the following function f(x) defines a probability mass
function.
x 0 1.5 4 4.8 5
f(x) 0.10 0.25 2m 0.41 m

Solution: 0.76+3m = 1. This gives, m = 0.08.

Problem 3: Find the value of m for which the following function f(x) defines a probability mass
function.
2𝑥+𝑚
𝑓(𝑥 ) = , x = 0, 1, 2, 3, 4.
25
Solution: f(0) = m/25, f(1) =(2+m)/25, f(2) = (4+m)/25, f(3) = (6+m)/25, f(4) = (8+m)/25.
20+5𝑚
f(0) + f(1) + f(2) + f(3) + f(4) = 1 implies = 1 or m = 1.
25

Problem 3 (3-22): An optical inspection system is to distinguish among different part types.
The probability of a correct classification of any part is 0.95. Suppose that three parts are

4
 Applied Statistics -Notes-2 Dr. T K Datta

inspected and that the classifications are independent. Let the random variable X denote the
number of parts that are correctly classified. Determine the probability mass function of X.
Solution: Let us define the events:
C: correct part specification
I: incorrect part specification
S = {CCC, CCI, CIC, CII, ICC, ICI, IIC, III}.
The possible values of X are 0, 1, 2, 3.
P(C) = 0.95 and P(I) = 1-0.95 = 0.05.
If f(x) defines the probability mass function of X, then
f(0) = 0.053 = 0.000125, f(1) = 3*0.95*0.052 = 0.007125,
f(2) = 3*0.952*0.05 = 0.135375, f(3) = 0.953 = 0.857375.
x 0 1 2 3
f(x) 0.000125 0.007125 0.135375 0.857375
This can also be defined by
𝑓(𝑥) = 𝐶𝑟3 × 0.95𝑟 × 0.053−𝑟 , 𝑟 = 0, 1, 2, 3.

Problem 4 (3-32): The distribution of the time until change (in days) of a Web site is
approximately in the following table.

Days until Changes Probability

1.5 0.15
3.0 0.15
4.5 0.45
5.0 0.20
7.0 0.05

Let the random variable X denote the days until change. Determine the pmf of the days until
change.

Solution: Let X denotes the random variable “days until change”

The pmf is:

x 1.5 3.0 4.5 5.0 7.0
f(x) 0.15 0.15 0.45 0.20 0.05

5
 Applied Statistics -Notes-2 Dr. T K Datta

Cumulative distribution function(CDF):

The cumulative distribution function (CDF) of random variable X is defined as

F(x)=P(X ≤ x), for all x ∈ R.
• 𝑃(𝑋 ≤ 𝑘) = 𝐹(𝑘)
• 𝑃(𝑚 < 𝑋 ≤ 𝑛) = 𝐹(𝑛) − 𝐹(𝑚),
• 𝑃(𝑋 > 𝑚) = 1 − 𝐹(𝑚).
CDF from PMF:
Example 1: Find the CDF of the random variable X whose pmf is given below:
x 1.5 3.0 4.5 5.0 7.0
f(x) 0.15 0.15 0.45 0.20 0.05
Use F(x) find 𝑃(𝑋 ≤ 4.5), 𝑃(𝑋 < 4.5), 𝑃(𝑋 ≥ 3.0), 𝑃(𝑋 > 4.5), 𝑃(3.0 < 𝑋 ≤ 7), 𝑃(𝑋 >
7), 𝑃(𝑋 < 1).
Solution: The cumulative distribution function F(x)of X is given below:
x 1.5 3.0 4.5 5.0 7.0
F(x) 0.15 0.30 0.75 0.95 1

𝑃(𝑋 ≤ 4.5) = 𝐹(4.5) = 0.75, 𝑃(𝑋 < 4.5) = 𝐹(3.0) = 0.30,

𝑃(𝑋 ≥ 3.0) = 1 − 𝐹(1.5) = 0.95, 𝑃(𝑋 > 4.5) = 1 − 𝐹(4.5) = 0.25,
𝑃(3.0 < 𝑋 ≤ 7) = 𝐹(7) − 𝐹(3) = 0.7, 𝑃(𝑋 > 7), 𝑃(𝑋 < 1).

Expectation, variance and standard deviation of a discrete random variable:

If the probability mass function of a discrete random variable X is f(x), then the mean or
expectation of the random variable is denoted by E(X) and is defined by
𝜇 = 𝐸(𝑋) = ∑𝑎𝑙𝑙 𝑥 𝑥𝑓(𝑥).
Variance is 𝜎 2 = 𝑉𝑎𝑟(𝑋) = ∑𝑎𝑙𝑙 𝑥(𝑥 − 𝜇)2 𝑓(𝑥) = ∑𝑎𝑙𝑙 𝑥 𝑥 2 𝑓(𝑥) − 𝜇 2 .

Standard deviation = 𝜎 = 𝑆𝐷(𝑋) = √𝜎 2 = √𝑉𝑎𝑟(𝑋).

Formula: 𝑬[𝒉(𝑿)] = ∑𝒂𝒍𝒍 𝒙 𝒉(𝒙)𝒇(𝒙).

Properties:
1. E(aX+b) = aE(X) + b.
2. Var(aX+b) = a2Var(X).
Example 1: The number of email messages received per hour has the following distribution:
x (number of messages): 10 11 12 13 14 15
f(x): 0.08 0.15 0.30 0.20 0.20 0.07
a) Find the cumulative distribution table.
b) Calculate E(X), var(X) and SD(X).

6
 Applied Statistics -Notes-2 Dr. T K Datta

[Ans:
a) x: 10 11 12 13 14 15 CDF table
f(x): 0.08 0.23 0.53 0.73 0.93 1
b) E(X) = 10*0.08 + 11*0.15+ 12*0.30 +13*0.20 +14*0.20 +15*0.07
= 0.8+1.65+3.6+2.6+2.8+1.05 = 12.5.
Var(X) = E(X2) – {E(X)}2

= [102*0.08 + 112*0.15+ 122*0.30 +132*0.20 +142*0.20 +152*0.07] – 12.52

= 1.85.
SD(X) = √1.85 = 1.36.]
Example 2: Errors in an experimental transmission channel are found when the transmissions
are checked by a certifier that detects missing pulses. The number of errors found in a eight-
bit byte is a random variable X with the following distribution:
F(x) = 0 when x<1
= 0.7 when 1 ≤x < 4
= 0.9 when 4≤ x < 7
= 1 when 7≤ x.
Determine each of the following probabilities:
(a) P(X ≤ 4) (b) P(X > 5) (c) P(X ≤5) (d) P(X > 7) (e) P(X ≤ 2).

[Solution: a) P(X ≤ 4) = F(4) = 0.9; b) P(X > 5) = 1- F(5) = 1 – 0.9 = 0.1;

(c) P(X ≤5) = F(5) = 0.9; (d) P(X > 7) = 1 – F(7) = 1 – 1 = 0;
(e) P(X ≤ 2) = F(2) = 0.7.]