You are on page 1of 12

NATIONAL DIPLOMA IN QUANTITY SURVEYING-

575/15/TN/0
SUBJECT TITLE: STATISTICS
SUBJECT CODE: 575/15/S06

INTRODUCTION TO PROBABILITY
What is a Statistical Experiment?
All statistical experiments have three things in common:

 The experiment can have more than one possible outcome.


 Each possible outcome can be specified in advance.
 The outcome of the experiment depends on chance.

A coin toss has all the attributes of a statistical experiment. There is more than one possible
outcome. We can specify each possible outcome (i.e., heads or tails) in advance. And there is
an element of chance, since the outcome is uncertain.

The Sample Space

 A sample space is a set of elements that represents all possible outcomes of a statistical
experiment.
 A sample point is an element of a sample space.
 An event is a subset of a sample space - one or more sample points.

Probability of an Event

With some statistical experiments, each sample point is equally likely to occur. In this
situation, the probability of an event is very easy to compute. It is:

𝑛(𝐸)
Number of sample points in event =𝑛(𝑆)
P(E) =
Number of sample points in sample space
Where 0 ≤ 𝑃(𝐸) ≤ 1 and ∑𝑛𝑖=1 𝑃(𝐸𝑖 ) = 1

Think about the toss of a single die. The sample space consists of six possible outcomes (1, 2,
3, 4, 5, and 6). And each outcome is equally likely to occur. Suppose we defined Event A to
be the die landing on an odd number. There are three odd numbers (1, 3, and 5). So, the
probability of Event A would be 3/6 or 0.5.

Types of events

 Two events are mutually exclusive if they have no sample points in common.

1
 Two events are independent when the occurrence of one does not affect the probability
of the occurrence of the other.
 Event – an occurrence in a defined context (can also be termed an outcome)
 Elementary event –is an event which cannot be broken down into sub events
 Compound event –an event which comprises two or more elementary events
 Sequential event –is an event that comprises a sequence of elementary events
 Trial –is a single attempt to obtain a defined event
 Complementary events –are events where one or other is certain to occur but not both
together

Test Your Understanding

1. Suppose I roll a die. Is that a statistical experiment?

Yes. Like a coin toss, rolling dice is a statistical experiment. There is more than one
possible outcome. We can specify each possible outcome in advance. And there is an
element of chance.

2. When you roll a single die, what is the sample space.

The sample space is all of the possible outcomes - an integer between 1 and 6.

3. Which of the following are sample points when you roll a die - 3, 6, and 9?

The numbers 3 and 6 are sample points, because they are in the sample space. The
number 9 is not a sample point, since it is outside the sample space; with one die, the
largest number that you can roll is 6.

4. Which of the following sets represent an event when you roll a die?

A. {1}
B. {2, 4,}
C. {2, 4, 6}
D. All of the above

The correct answer is D. Remember that an event is a subset of a sample space. The
sample space is any integer from 1 to 6. Each of the sets shown above is a subset of the
sample space, so each represents an event.

5. Consider the events listed below. Which are mutually exclusive?

A. {1}
B. {2, 4,}
C. {2, 4, 6}

Two events are mutually exclusive, if they have no sample points in common. Events A

2
and B are mutually exclusive, and Events A and C are mutually exclusive; since they
have no points in common. Events B and C have common sample points, so they are not
mutually exclusive.

6. Suppose you roll a die two times. Is each roll of the die an independent event?

Yes. Two events are independent when the occurrence of one has no effect on the
probability of the occurrence of the other. Neither roll of the die affects the outcome of
the other roll; so each roll of the die is independent.

PROBABILITY SYMBOLS

a) P(. . . ) – this is the symbol for the probability of whatever event is stated between the
brackets ie. Tossing a coin P(head) = ½, P(tail)=1/2 as well
b) 𝑃(𝐴 ∩ 𝐵) − 𝑠𝑦𝑚𝑏𝑜𝑙 ∩ 𝑖𝑠 𝑟𝑒𝑎𝑑 𝑎𝑠 𝐴𝑁𝐷, 𝑠𝑜 𝑃(𝐴 ∩
𝐵)𝑖𝑠 𝑡ℎ𝑒 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑏𝑜𝑡ℎ 𝑒𝑣𝑒𝑛𝑡𝑠 𝐴 𝑎𝑛𝑑 𝐵 (𝑖𝑛𝑡𝑒𝑟𝑠𝑒𝑐𝑡𝑖𝑜𝑛)
c) 𝑃(𝐴 ∪ 𝐵) −∪ 𝑖𝑠 𝑟𝑒𝑎𝑑 𝑎𝑠 𝑂𝑅 𝑠𝑜 𝑃(𝐴 ∪ 𝐵)𝑖𝑠 𝑡ℎ𝑒 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝐴 𝑜𝑟 𝐵 (𝑢𝑛𝑖𝑜𝑛)
d) 𝑃(𝐴̅). 𝑇ℎ𝑒 𝑏𝑎𝑟 𝑜𝑣𝑒𝑟 𝑡ℎ𝑒 𝑠𝑦𝑚𝑏𝑜𝑙 𝑓𝑜𝑟 𝑎𝑛 𝑒𝑣𝑒𝑛𝑡 𝑖𝑛𝑑𝑖𝑐𝑎𝑡𝑒𝑠 𝑡ℎ𝑎𝑡 𝑖𝑡 𝑤𝑖𝑙𝑙 𝑛𝑜𝑡 𝑜𝑐𝑐𝑢𝑟.
𝑆𝑜𝑚𝑒𝑡𝑖𝑚𝑒𝑠 𝑃(𝐴̀)𝑖𝑠 𝑢𝑠𝑒𝑑 𝑓𝑜𝑟𝑃(𝐴̅)

LAWS OF PROBABILITY

1. The addition law of probability (union of events /addition rule) OR rule


This rule states that the total probability of either of the two mutually exclusive
events occurring is the sum of the probabilities of each of the events occurring ie.
𝑃(𝐴 ∪ 𝐵) = 𝑃(𝐴) + 𝑃(B)
Example
Workout the probability of throwing a one or a six with a die
Answer
1 1 2 1
+6=6=3
6
Exercise
Workout the probability of obtaining a score greater than 2 when a die is rolled.
2. The general addition rule
When A and B are not mutually exclusive we apply the general rule for the addition
of probabilities
-It applies when we want to find the probability of event A or event B occurring,
including the probability that both occur simultaneously
𝑃(𝐴 ∪ 𝐵) = 𝑃(𝐴) + 𝑃(𝐵) − 𝑃(𝐴 ∩ 𝐵)

Example

The head of department has to select a new trainee from an intake of 20 trainees joining a
manufacturing company. There are 8 female trainees, 3 of the females and 3 of the males have
Journeyman class 1. The head of department would like a female trainee, preferably with
Journeyman class 1. However, it is decided to put the names of all trainees into a hat and to draw a
name at random so that there is an equal chance of selecting any of the trainees. What is the
probability that either a female trainee or one with Journeyman class 1 will be selected?

3
Answer

𝑃(𝑠𝑒𝑙𝑒𝑐𝑡𝑖𝑛𝑔 𝑎 𝑓𝑒𝑚𝑎𝑙𝑒 𝑡𝑟𝑎𝑖𝑛𝑒𝑒 𝑜𝑟 𝑡𝑟𝑎𝑖𝑛𝑒𝑒 𝑤𝑖𝑡ℎ 𝑎 𝐽𝑜𝑢𝑟𝑛𝑒𝑦𝑚𝑎𝑛 𝑐𝑙𝑎𝑠𝑠 1)


= 𝑃(𝑠𝑒𝑙𝑒𝑐𝑡𝑖𝑛𝑔 𝑎 𝑓𝑒𝑚𝑎𝑙𝑒) + 𝑃(𝑠𝑒𝑙𝑒𝑐𝑡𝑖𝑛𝑔 𝑎 𝑡𝑟𝑎𝑖𝑛𝑒𝑒 𝑤𝑖𝑡ℎ 𝑗𝑜𝑢𝑟𝑛𝑒𝑦𝑚𝑎𝑛 𝑐𝑙𝑎𝑠𝑠 1)
− 𝑃(𝑓𝑒𝑚𝑎𝑙𝑒 ∩ 𝑗𝑜𝑢𝑟𝑛𝑒𝑦𝑚𝑎𝑛 𝑐𝑙𝑎𝑠𝑠 1 ℎ𝑜𝑙𝑑𝑒𝑟)
8 6 3
= + −
20 20 20
11
=
20
3. The multiplication law of probability (the AND rule)
It states that the probability of a sequential event is the product of all its elementary
events ie. When we toss a coin into the air twice, it must come down each time and
it can fall in any one of four equally likely ways, H&H, H&T, T&H, T&T, now
𝑝(𝑜𝑏𝑡𝑎𝑖𝑛𝑖𝑛𝑔 2 ℎ𝑒𝑎𝑑𝑠)
= 𝑝(ℎ𝑒𝑎𝑑 𝑜𝑛 𝑓𝑖𝑟𝑠𝑡 𝑡ℎ𝑟𝑜𝑤) × 𝑃(𝐻𝑒𝑎𝑑 𝑜𝑛 𝑠𝑒𝑐𝑜𝑛𝑑 𝑡ℎ𝑟𝑜𝑤)
1 1 1
=2×2=4

It applies when events are independent, when the probability of either of them occurring is not
affected by occurrence or non-occurrence of the other

Example

Out of a crate of 40 bottles, 2 are under filled. An inspector selects one bottle at random from the
crate. What is the probability that it is not under filled? In a second crate of 40 bottles , 4 are under
filled. The inspector randomly selects a bottle from this crate. What is the probability that both
bottles are not under filled?

Answer
2 1 19
1st crate 𝑃(𝑢𝑛𝑑𝑒𝑟 𝑓𝑖𝑙𝑙𝑒𝑑) = 40 = 20 , 𝑃(𝑛𝑜𝑡 𝑢𝑛𝑑𝑒𝑟𝑓𝑖𝑙𝑙𝑒𝑑) = 20,
4 1 9
2nd crate 𝑃(𝑢𝑛𝑑𝑒𝑟𝑓𝑖𝑙𝑙𝑒𝑑) = 40 = 10 , 𝑃(𝑛𝑜𝑡 𝑢𝑛𝑑𝑒𝑟𝑓𝑖𝑙𝑙𝑒𝑑) = 10,
19 9 171
So 𝑃(𝑏𝑜𝑡ℎ 𝑏𝑜𝑡𝑡𝑙𝑒𝑠 𝑛𝑜𝑡 𝑢𝑛𝑑𝑒𝑟𝑓𝑖𝑙𝑙𝑒𝑑) = × =
20 10 200

4. The General multiplication rule


It applies when events are not independent

Example

From our earlier example, the head of department selects the names of two trainees at random
from the hat. What is the probability that two females are selected, assuming that the first name is
not replaced before selecting the second?

Answer
8 7 8 7 14
1st selection is , 2nd selection is , therefore × =
20 19 20 19 95

Illustration on a tree diagram

4
Conditional Probability

When the probability of an event depends upon the conditions surrounding it we have conditional
probability

𝑃(𝐴⁄𝐵)
Example

A class contains 6 males and 4 females. Find the probability that a male student will take 1st prize
and a female 2nd prize?

Answer
6 4 4
× =
10 9 15

BAYES THEOREM

Given that a particular event S has actually occurred, then the probability that a given possible prior
event E actually occurred is
𝑃(𝐸) × 𝑃(𝑆|𝐸)
𝑃(𝐸|𝑆) =
𝑃(𝑆|𝑎𝑛𝑦 𝑝𝑟𝑖𝑜𝑟 𝑒𝑣𝑒𝑛𝑡)
𝑃(𝐸) × 𝑃(𝑆|𝐸)
= , 𝑤ℎ𝑖𝑐ℎ 𝑖𝑠 𝑐𝑎𝑙𝑙𝑒𝑑 𝑡ℎ𝑒 𝐵𝑎𝑦𝑒𝑠 𝑇ℎ𝑒𝑜𝑟𝑒𝑚/𝑙𝑎𝑤
∑𝑛𝑖=1[ 𝑃(𝐸𝑖 )
× 𝑃(𝑆|𝐸𝑖 )]

Example

95% of workers in a plant wear safety attire, 64% of workers involved in serious accidents die if not
wearing safety attire where as 12% of those that wear safety attire die. Calculate correct to
3significant figures the percentage of workers involved in serious accidents who died and were not
wearing safety attire.

Answer

Let D be the event ‘died as a result of serious accident’ and S be the event ‘wears safety attire’

Tree diagram

𝑃(𝑆̅)×𝑃(𝐷 |𝑆̅)
We require 𝑃(𝑆̅|𝐷) =
𝑃(𝐷 |𝑆̅)×𝑃(𝑆̅)+𝑃(𝐷 |𝑆)×𝑃(𝑆)

0.64 × 0.05
=
0.64 × 0.05 + 0.12 × 0.95
= 0.2192

5
= 21.9%
EXPECTATION AND VARIANCE
Expectation
If all the events in a probability situation have values, then it becomes possible to compute
what the average value (mean) would be if the situation repeated itself frequently.
Discrete probability distribution
Properties
-any probability distribution for a discrete random variable has the following properties
1. 0 ≤ 𝑃(𝑥 = 𝑥𝑖 ) ≤ 1, 𝑓𝑜𝑟 𝑖 = 1, 2, 3, . . . 𝑛
2. ∑𝑛𝑖=1 𝑃(𝑥𝑖 ) = 1
Probability distribution of a discrete random variable (r.v)
- Let 𝑥 be a discrete r.v , the probability distribution of 𝑥 describes how the
probabilities are distributed over the possible values of 𝑥
- The probability that the r.v. will assume the value 𝑥𝑖 is denoted by 𝑃(𝑥 = 𝑥𝑖 )

Example
3 games were played and the following results were obtained
𝑃(𝑤𝑖𝑛𝑛𝑖𝑛𝑔 0 𝑔𝑎𝑚𝑒𝑠) = 0.65,
𝑃(𝑤𝑖𝑛𝑛𝑖𝑛𝑔 1 𝑔𝑎𝑚𝑒) = 0.15,
𝑃(𝑤𝑖𝑛𝑛𝑖𝑛𝑔 2 𝑔𝑎𝑚𝑒𝑠) = 0.10
𝑎𝑛𝑑 𝑃(𝑤𝑖𝑛𝑛𝑖𝑛𝑔 3 𝑔𝑎𝑚𝑒𝑠 = 0.10
Construct a probability distribution table for the data.
Answer
Probability distribution table

𝑥 0 1 2 3
𝑃(𝑥 = 𝑥𝑖 ) 0.65 0.15 0.1 0.1

Expected value of a discrete random variable


Mean value of a r.v. in many trials 𝐸(𝑥)
It is defined to be the quantity

6
𝑛

𝐸 (𝑥 ) = ∑ 𝑥𝑖 𝑃(𝑥𝑖 ), 𝑤ℎ𝑒𝑟𝑒 𝑃(𝑥𝑖 ) = 𝑃(𝑥 = 𝑥𝑖 )


𝑖=1
Example
From the probability distribution table above, find 𝐸(𝑥)
Answer
𝑛

𝐸(𝑥) = ∑ 𝑥𝑖 𝑃(𝑥𝑖 ) = 0 × 0.65 + 1 × 0.15 + 2 × 0.10 + 3 × 0.10 = 0.65


𝑖=1

Variance of a discrete random variable


The variance of a probability distribution is represented by an expected value in the same
manner as mean
𝑛

𝑉𝑎𝑟(𝑥 ) = ∑(𝑥𝑖 )2 𝑃(𝑥𝑖 ) − [𝐸 (𝑥 )]2


𝑖=1
Example
Find the variance for the data above
Answer
𝑛

𝑉𝑎𝑟(𝑥) = ∑(𝑥𝑖 )2 𝑃(𝑥𝑖 ) − [𝐸(𝑥)]2


𝑖=1

= 02 × 0.65 + 12 × 0.15 + 22 × 0.10 + 32 × 0.10 − 0.652


= 1.03

𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = √1.03 = 1.014


Exercise
The probability distribution of a random variable x is as shown in the table below

𝑥 1 2 3 4 5
𝑃(𝑥 = 𝑥𝑖 ) 0.1 0.3 𝑦 0.2 0.1

Find
a) The value of y
b) 𝐸(𝑥)
c) 𝑉𝑎𝑟(𝑥)

7
Expected value of a continuous Random Variable(r.v)
+∞
𝐸(𝑥) = ∫ 𝑥𝑓(𝑥)𝑑𝑥
−∞

Example
12𝑥(1 − 𝑥)2 , 0 ≤ 𝑥 ≤ 1
Given 𝑓(𝑥) = { , find E(x)
0 , 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒
Soln
1 1
12 5
𝐸(𝑥) = ∫ 𝑥𝑓(𝑥)𝑑𝑥 = ∫ 12𝑥 2 (1 − 𝑥)2 𝑑𝑥 = 4𝑥 2 − 6𝑥 4 + 𝑥 = 0.4
0 0 5
Variance of a continuous r.v
+∞
𝑉𝑎𝑟(𝑥) = ∫ 𝑥 2 𝑓(𝑥)𝑑𝑥 − [𝐸(𝑥)]2
−∞

For our above example 𝑣𝑎𝑟(𝑥) = 0.04


Exercise
𝑥
; 0≤𝑥≤1
2
a) Verify that the function 𝑓(𝑥) = {1 ; 1 ≤ 𝑥 ≤ 2.5 can represent a probability
2
0; 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
density function
b) Find the mean and the variance of the function in (a) above giving your answers to
3s.f
Soln
a) Simply show that the total area under the curve is 1
1𝑥 2.5 1
i.e ∫0 2 𝑑𝑥 + ∫1 𝑑𝑥 = 1
2
1 𝑥 2.5 1
b) Mean = 𝐸(𝑥) = ∫0 𝑥. 𝑑𝑥 + ∫1 𝑥. 𝑑𝑥 = 1.48
2 2
1 2.5
𝑥 1
𝑉𝑎𝑟(𝑥) = [∫ 𝑥 . 𝑑𝑥 + ∫ 𝑥 2 . 𝑑𝑥] − 1.482 = 0.374
2
0 2 1 2

8
Binomial Probability Distribution (discrete distribution)
In a Binomial distribution you will be knowing the number of trials
It has only two outcomes that
𝑃(𝑠𝑢𝑐𝑐𝑒𝑠𝑠) = 𝑝
𝑎𝑛𝑑 𝑃(𝑓𝑎𝑖𝑙𝑢𝑟𝑒) = 1 − 𝑝 = 𝑞
𝑛
𝑃(𝑥 = 𝑥𝑖 ) = ( ) 𝑝 𝑥 𝑞 𝑛−𝑥
𝑥
Conditions under which a Binomial distribution can be applied
-applies when we have only two possible different characteristics in a population
-the validity of the formula depends very much on the fact that to all intends and purposes p
remains virtually unchanged during the entire selection of a sample
-the trials are independent
-the outcome of each trial is deemed either a success or a failure
-a finite number , n, trials are carried out
Example
A fair coin is tossed 5 times, find
a) The probability of obtaining 3 heads
b) The probability of obtaining at least 3 heads
Solution
13 15−3 5
a) 𝑃(𝑥 = 3) = (53) 2 = 16 = 0.31
2

b) 𝑃(𝑥 ≥ 3) = 𝑃(𝑥 = 3) + 𝑃(𝑥 = 4) + 𝑃(𝑥 = 5)

5 13 12 5 14 11 5 15 10
=( ) +( ) +( )
3 2 2 4 2 2 5 2 2

5 5 1
= + +
16 32 32
1
=
2
= 0.5
Mean of a Binomial distribution
𝐸(𝑥) = 𝑛𝑝
Example

9
From the example above,
1
𝐸(𝑥) = 𝑛𝑝 = 5 × = 2.5
2

Variance of a Binomial distribution


𝑉𝑎𝑟(𝑥) = 𝑛𝑝𝑞

𝑆𝐷 = √𝑛𝑝𝑞

Example
From above
1 1
𝑉𝑎𝑟(𝑥) = 𝑛𝑝𝑞 = 5 × × = 1.25
2 2
Exercise
1. A construction company buying construction materials has 8 applications a week for
1
estimates. The probability that any quote will be accepted is 6
a) Calculate the probability that 𝑥 quotes will be accepted for values of 𝑥 from 0 to 8
b) Use your answer to part (a) to find
i) The probability that more than half of the estimates will be accepted
ii) The most likely number of new estimates accepted each week
c) Find the standard deviation of the distribution
Soln Hint
a) Draw a probability distribution table for the stated range
b) (i) 0.0046, (ii)the one with the highest probability
c) 1.05

10
Assignment
Question 1
a) An underground mine has 5 pumps installed for pumping out storm water, the probability
1
of any one of the pumps failing during the storm is . What is the probability that
8
i) At least 2 pumps will be working [3]
ii) All the pumps will be working during a particular storm [2]
c) A die having 4 red and 2 yellow faces is thrown twice. For each red face showing the thrower
is paid 9cents and for each yellow face 27 cents. What is the thrower’s expectation? [5]

Question 2

a) The probability of a component failing in one year due to excessive temperature is


1 1 1
20
, 𝑑𝑢𝑒 𝑡𝑜 𝑒𝑥𝑐𝑒𝑠𝑠𝑖𝑣𝑒 𝑣𝑖𝑏𝑟𝑎𝑡𝑖𝑜𝑛 𝑖𝑠 25
𝑎𝑛𝑑 𝑑𝑢𝑒 𝑡𝑜 𝑒𝑥𝑐𝑒𝑠𝑠𝑖𝑣𝑒 ℎ𝑢𝑚𝑖𝑑𝑖𝑡𝑦 𝑖𝑠 50
Determine
the probabilities that during a one year period, a component:
i) Fails due to excessive temperature and excessive humidity [3]
ii) Fails due to excessive vibration or excessive humidity [3]
iii) Will not fail because of both excessive temperature and excessive humidity [4]

Question 3

a) In a certain crime-ridden area, the probability that a house will be burgled on a particular
night is 0.03. A burglar alarm is installed in a house in the area. If the house is burgled, the
probability that the alarm will go off is 0.81. If the house is not burgled, the probability that
it will go off by accident is 0.1. Find the probability that the house has been burgled, if the
alarm has gone off. [5]
b) What is the probability of throwing a 3 and a 6 with two throws of a die? [3]
c) Define the term mutually exclusive events [2]

Question 4

a) The probability that component A will operate satisfactorily for 5 years is 0.8 and that B will
operate satisfactorily over the same period is 0.75. Find the probability that in 5 years
i) Both components operate satisfactorily [2]
ii) Only component A will operate satisfactorily [2]
iii) Only component B will operate satisfactorily [2]
b) Concrete blocks are tested and it is found that on average, 7% fail to meet the required
specification. For a batch of 9 blocks, determine the probability that
i) 3 blocks will fail to meet the specification [2]
ii) Less than 4 blocks will fail to meet the specification [2]

………………..OUT OF 40…………………………………………………..

11
12

You might also like