Probability and its application

By

Dr. Md. Manir Hossain Mollah

Probability

FOCUS POINTS
 Introduction
 Why do we need probability?
 What is trial/experiment and event?
 Define probability.
 Properties and rules of probability
 Compute probability of events.
Introduction

People use the term probability many times in our daily life

For example:

A doctor says

―A patient has a 50-50 chance of surviving an operation.

―95% certain that he has a lung cancer

―nine out of ten patients take drugs regularly (researcher says)

i.e., probability is the foundation of statistical inference.
Cont’d…

Probability – expressed in term of fractions (Mathematically)

Probability – generally expressed in term of percentage

(multiplying the fractions by 100)

Probability of occurrence – between zero (0) and one (1)

Why do we need probability?
Statistical studies usually are not based on a total
population survey but are based on a random sample of
the total population.
A random sample is a random selection of observational
units from the population
A simple random sample is selected so that each unit has
the same probability to get sampled.
The resulting sample should be representative for the
population.
Cont’d…

 To draw conclusions from the information of a sample

on characteristics of the entire population, we need the
theory of probabilities.
 Probabilities allow us to quantify and compare the
frequency with which we observe a certain event (e.g.
the occurrence of death or an illness)
 Probabilities are the basis of estimation and statistical
testing
What is trial/experiment and event?
Trial/Experiment:
Any planned activity or process that result in some
definite outcome is called Trial or Experiment.
Event:
Any outcome from a trail is called Event.
eg.
(1) Coin tossing is a trial or experiment
‘Head’ or ’Tail’ – outcome/event
(3) Expanded program on Immunization (EPI) - trial.
Disease ‘occurred’, ‘Not Occurred’ - event.
(2) Pregnancy – trial/experiment
‘Miscarriage’, ‘Live birth’, ‘Still birth’ - event
Characteristics of events
1. Mutually exclusive
One and only one outcome can happen out of a single
trial.
- If one outcome happens than other outcome(s) will
not happen
or
In other words two or more outcomes cannot happen
at a time.

Mathematically, two events A and

B can be expressed as Mutually
exclusive if P(A and B) = 0
Characteristics of events

Example:
Let us consider, a coin is tossed. There are two possible
outcomes: Head and Tail. In any toss (trial) it will be
either Head or Tail. Both Head and Tail cannot happen
simultaneously.
Characteristics of events
1. Exhaustive
All possible outcomes together is called “Exhaustive”.
Example 2: In the experiment of tossing a coin:
X = event of getting a "tail“
Y= event of getting a "head“
Solution:
The event of flipping two coins.
M = the events of at least one head = {HH, HT, TH} and
N = the events of at least one tail = {TH, TT, HT}.
- If we are considering only event of at least one head it
cannot be exhaustive as it is not certain to happen
- But when we take both events together they are certain
to happen for any one outcome so the events together are
exhaustive.
Characteristics of events
3. Equally likely
If probability of getting Head(H) and Tail(T) is equal , then H and T are
called equally likely event.
Example : For a coin toss P(H)=P(T)=0.5; if a coin is unbiased.

An example of not Equally likely event

Select three children at random and our interested in the probability
that none of the children have any birth defects. D to represent a
child was born with a birth defect and N to represent the child born
without birth defect. Thus
S = {DDD, NDD, DND, DDN, DNN, NDN, NND, NNN}
Are the events DDD (all three children are born with birth defects)
and NNN (none of the children are born with birth defects) equally
likely?
It should be reasonable to you that P(NNN) is much larger than
P(DDD). This is because P(N) and P(D) are not equally likely events. It
is rare (certainly not 50%) for a randomly selected child to be born
with a birth defect.
 Definition

Hopefully it is clear that these two outcomes are not equally likely!!
Let’s consider a more common example.
EXAMPLE: Birth Defects
Suppose we randomly select three children and we are interested in the probability
that none of the children have any birth defects.
We use the notation D to represent a child was born with a birth defect and N to
represent the child born with NO birth defect. We can list the possible outcomes just
as we did for the coin toss, they are:
{DDD, NDD, DND, DDN, DNN, NDN, NND, NNN}
Are the events DDD (all three children are born with birth defects) and NNN (none of
the children are born with birth defects) equally likely?
It should be reasonable to you that P(NNN) is much larger than P(DDD).
This is because P(N) and P(D) are not equally likely events.
It is rare (certainly not 50%) for a randomly selected child to be born with a birth
defect.
Characteristics of events
4. Independence of Events
If the occurrence of an event is not affected by the occurrence
of another event, then events are said to be independent.
i.e., mathematically, P(E1 and E2)= P(E1) * P(E2)
* Sometimes we can just use common sense to guide us as to
whether two events are independent.
EXAMPLE:
Two people are selected simultaneously and at random from all
people in the Dhaka.
Let B1 be the event that one of the people has blue eyes and B2
be the event that the other person has blue eyes.
In this case, since they were chosen at random, whether one of
them has blue eyes has no effect on the likelihood that the other
one has blue eyes, and therefore B1 and B2 are independent.
What is sample space?
Sample space : All the possible outcomes of a trail
For example:
 Experiment of tossing a fair coin. The sample space of
the experiment is

S = {Head, Tail} or simply {H, T}

 Experiment of tossing three fair coins. The sample

space of the experiment is

S = {HHH, HHT, HTH, THH, HTT, THT, TTH, TTT}

Definition of probability
In general, there are two types of probability
1. Theoretical/Classical/Equally-likely Approach:
Let n(E) be the no. of favorable outcomes of an event E and
n(S) be the total no. of possible outcomes, all of which
are mutually exclusive and equally likely, in a trail, then
the probability is defined as
Example

 Probability of rolling (1,1) with two dice is

P(1,1) = 1/36 = 0.028 = 2.8%
 Probability of 7 or 11 from rolling two dice is
P(7 or 11) = (6+2)/36 = 0.222
 What is the probability of rolling (3, 3) with two dice?
 What is the probability of double (i, i) with two dice?
Another example: Experiment on patient

A doctor X during his career, he

mad 50 successful surgery out of
60 cancer patients.
E = He did a successful surgery
-What is the P(E)?
P(E) = 50/60 = 0.833
Definition of probability

2. Experimental Probability/Relative Frequency Probability:

It is defined as the ratio of trails where the desired event
occurs divided by the no. of trails.
*Usually observed over a long run

# of times E occurs
P( E ) 
# of trail repeated
Example
Let a researcher collected the age of each child admitted
in child care hospital. The following histogram summarizes
the data he collected.

# of child admitted in child care

hospital of age between10 and 15
P( E ) 
# of trail
3 3
=   0.3
3  2  3  1  1 10

 Based on this data, what is a reasonable estimate of the probability

that the next person to enter child care hospital is between 10 and
15 years olds
Properties of probability

 The probability ranges between 0 and 1

(i.e., 0 ≤ P(event) ≤ 1)

 If an outcome is sure, it has a probability of 1

(i.e., P(Certain event) = 1) Having one Tuesday in this week?
Each week has a Tuesday, so probability = 1.
 If an outcome cannot occur, its probability is 0
(i.e., P(impossible event) = 0)
Ex. Having two Tuesday in this week? Each week has only one Tuesday, so
probability = 0
Rules of probability
There are some important rules of probability
1. Multiplication rule
Let A and B be two events in a sample space S
A. If both are independent, then
P(A n B) or P(A & B)=P (A)×P(B) = P(B) × P(A)
or
P(B|A) = P(B)
or
(1-P(A))×(1-P(B))=1-(P(A)+P(B)-P(A n B))
= 1-P(A U B)
B. If both are dependent, then
P(A & B) = P (A) × P (B/A) or P(B) × P(A/B)
where P(A) and P(B) are called Unconditional or marginal
probability P(B/A) is Conditional probability
P(A & B) is the joint probability (note: Joint Prob.= Marginal Prob.
× Conditional Prob.)
 Example

Independence: P(TP & DP) = P (TP) × P (DP) = P(DP)

× P(TP)
Test of Independence; LHS: P(TP & DP) = 3/36 = 1/12
RHS = P (TP) × P (DP) = 6/36 × 18/36 = 1/6 × 1/2 = 1/12
Or
P(DP|TP) = 3/6 = 1/2 = P(DP) = 18/36 = 1/2
Since LHS = RHS, thus both events are independent
Thus joint probability: P(TP and DP) = 1/12 = 0.083
This means the joint probability that participant is female and responded
 Example

Dependence: P(TP & DP) = P(TP) × P(DP|TP)

Test of dependence; RHS = 6/36 × 3/6 = 1/6 × ½=1/12
Since LHS ≠ RHS, they are not dependent each other
Rules of probability
There are some important rules of probability
1. Addition Rule (mutually exclusive events)
Let us consider, A and B are two mutually
exclusive events, than the probability that
either A or B will occur is,
P(A or B) = P (A) + P (B)
i.e.,

A B
S
Example
Table: OPD attendance by disease at a selected
hospital
Disease No. of patients
Diabetes 50
Heart Disease 30
Respiratory 20
others 50
Total 150
P(Diab or Heart Dis ) = P(D 0r HD) = P(D) + P(HD)
= 50/150 + 30/150
= 0. 53
2. Addition Rule (mutually non exclusive events)
Let us consider, A and B are two mutually non
exclusive events, than the probability that
either A or B will occur is,
P(A or B) = P (A) + P (B) – P (A and B)
i.e.,

A B

S
Example

The probability that a student passes Biostatistics is

2/3 and he passes epidemiology is 4/9.If the
probability of passing both courses is 4/5, what is the
probability that, he will pass at least one courses?

Solution
P(B or E)= P(B)+ P(E)- P(B and E)
= 2/3 + 4/9 - 4/5
=14/45
=0.311
Question
A medical trial into the effectiveness of a new medication was carried out. 120 females and
90 males took part in the trial. Out of those people, 50 females and 30 males responded
positively to the medication. Given below is a contingency table with the given information
filled in.
1.What is the probability that the medicine gives a positive result for females?
2.What is the probability that the medicine gives a negative result for males?
3.Was the medication's success independent of gender? Explain.

Female Male Totals

Positive 50 30 80
Negative 70 60 130
Totals 120 90 210
Soln.
The way the first question is phrased, we need to determine the probability that a person
responds positively if she is female.
This means that we do not include males in this calculation.
So, the probability that the medicine gives a positive result for females is the ratio between
the number of females who got a positive response and the total number of females. Thus
P(positive if female)=n(positive and female)/n(female)=50/120=5/12
Similarly, the probability that the medicine gives a negative result for males is:
P(negative if male)=n(negative and male)/n(male)=60/90=2/3

Independence test
We need to determine whether the effect of the medicine and the gender of a participant are
dependent or independent. According to the definition, two events are independent if and
only if
P(A and B)=P(A)×P(B)
We will look at the events that a participant is female and that the participant responded
positively to the trial.
P(female)=n(female)/n(total trials)=120/210=4/7
P(positive)=n(positive)/n(total trials)=80/210=8/21
P(female and positive)=n(female and positive)/n(total trials)=50/210=5/21
From these probabilities we can see that
P(female and positive)≠P(female)×P(positive)
and therefore the gender of a participant and the outcome of a trial are dependent events.