Lecture 5 Probability
Lecture 5 Probability
THEORY OF PROBABILITY
Introduction
In the theory of probability we deal with events which are the outcomes of
experiments, repeated under essentially homogeneous and similar conditions. An
experiment is defined as an action, which we conceive, do or intend to do. The experiments
are of two types:
(a) Deterministic experiments: Here chance does not play any part or role in the
outcome of an experiment. In other words, the given conditions completely determine the
outcome. That is to say, we can completely predict the outcome.
Thus for example, the pressure p and volume v of a gas are known to be related by
an equation of the form
pvk = c
from which we can determine p as soon as we know v; k and c being constants.
(b) Probabilistic experiments: Here chance plays an important role in the outcome
of an experiment. In other words, the given conditions do not determine completely the
outcomes and the phenomena is non-deterministic or probabilistic or stochastic.
(i) One cannot guess with certainty whether it will be a rainy day or a sunny day
tomorrow.
(ii) In rolling an honest or unbiased die, one cannot predict with certainty which
face will turn up.
An event
If a die is tossed, we may be interested in the "event", 'A', that the number on the top
face is an even number. The event corresponds to the outcomes 2, 4 and 6. If an event is
denoted by A, the complementary event, i.e., non-occurence of A is denoted by A. In the
present case, the complementary event is the occurrence of 1, 3 and 5. In general, if there
are n cases out of which m are favourable to A, n-m outcomes are favourable to A.
Probability
Outcomes: The results of an experiment are known as out comes. For example, in
the tetrazolium test there are two possible outcomes: stained and non-stained.
The porbability that an event A occurs is a measure of the level of certainty that the
outcome of a trial is A. If an experiment can result in n exhaustive, mutually exclusive and
equally likely outcomes, and m of these outcomes are favourable to an event A, then the
probability of the event A is defined as
If m = 0 then P(A) = 0 which means that A is impossible. If m = n then P(A) = 1 and the
event A is certain.
In general P(A) lies between these two extremes and a high value of P(A) indicates a high
level of certainty for A.
Example 1a
There are 10 chickens in a pen of which 5 are red, 3 are black and 2 are white. When the
door of the pen is opened the chickens are equally likely to escape in any order.
R R R R R
B B B W W
The probabilities that the first chicken out is red, black or white are 0.5, 0.3 and 0.2
respectively.
Other examples:
Exclusiveness
Example 1b
The chicken example gives a simple application of exclusiveness. The first chicken
out can either be red or black but cannot be both colours at the same time. The probability
that the first chicken out is either red or black is
Similarly the probability that the first chicken is either red or black or white is
This means that it is certain that the first chicken out is either red or black or white,
which is the expected solution.
Non-exclusiveness
If there is an overlap of two events then the sum of their probabilities does not give
the probability that either one or the other occurs since the overlap would be double
counted. An adjustment overcomes the problem.
Example 1c
In the chicken example suppose that all the black birds are male, all the white birds
are female, but of the red birds there are three males and two females.
RM RM RM RF RF
BM BM BM WF WF
Suppose that the problems is to find the probability that the first bird out is either red
or male.
Now P(R) = 0.5 and P(M) = 0.6 so notice that the sum of these probabilities exceeds
1, which cannot be true. The explanation is that the two events can overlap since it is
possible for a chicken to be both red and male at the same time. The solution is
Observe that the number of birds which are either red or male or both is 8, so the
problem could have been solved directly from the definition.
Independence
Two trials are independent if the outcome of the first trial has no effect on the
outcome of the second.
Suppose that two independent trials are conducted, then the probability that the
outcome of the first trial is A and the outcome of the second is B is given by the product
Example 1d
In the chicken example suppose that the door of the pen is opened so the chickens
come out, that they are rounded up then let out a second time. There are nine possible
arrangements of the coat colour for the first chicken out on the two occasions.
Total = 1.00
The nine possibilities are exclusive because none of them can occur simultaneously,
for example the sequence RB cannot occur if RW occurs, and RB is different from BR.
Moreover the list is exhaustive since all possibilities are covered. For any set of events that
is exclusive and exhaustive the total probability is 1.
As another example suppose it was required to calculate the probability that for the
same conditions
(a) a black chicken emerged first on either run.
(b) a black chicken emerged first both times.
(c) a black chicken did not emerge first on either occasion.
(d) a black chicken emerged first on exactly one run.
Solutions:
(a) The possibilities, which fulfil these conditions, are RB, BR, BB, BW, and WB. From
the table the sum of these probabilities is 0.51.
(b) The only possibility is BB, whose probability is 0.09
(c) The possibilities are RR, RW, WR, WW whose total probability is 0.49.
(d) Possibilities are BR, BW, RB, WB. Total probability 0.42.
Observe that the total probability for (a) and (c) is 1 since the two cases are
exclusive and exhaustive. Also note that the same point applies to (b), (c) and (d), as the
number of times that a black bird can emerge first in two runs is 0, 1 or 2.
Dependence
If there is dependence between two trials then the probability that the outcome are
respectively A and B is given by the product
where means the probability that B occurs given that A has already occurred.
Example 1e
In the chicken example there were 5 red, 3 black and 2 white birds and in example
1a they were released so that birds emerged in random order.
(a) Calculate the probability that the first out on release is red and that the second
out is black.
Solution:
Initially there are 5 red birds out of ten so the probability that the first out is red is
0.5. Now if the first out is taken to be red this leaves 4 red, 3 black and 2 white birds, so the
probability that the second bird is black given that the first is red is 3/9 or 1/3. The
probability P(RB) is given by
Example 1f
An urn contains 8 white and 3 red balls. If two balls are drawn at random, find the
probability that (a) both are white, (b) both are red and (c) one is of each colour.
Solution:
Total number of balls in the urn = 8 + 3 = 11
(c) The number of favourable cases for drawing one white ball and one red ball is
8
C1 x 3C1 = 8 x 3 = 24.
Empirical Approach
Professor Dr. Khandoker Saif Uddin Lecture # 5, Page 7
Theory of Probability
Thus
If an event A occurs r times in a series of n independent trials, all made under the
same essential conditions, the ratio r/n is called the relative frequency of the event A in n
trials. The limit of , as n tends to infinity, is the probability P(A) of the event A. That is
Subjective Probability
Joint Probability:
A probability that measures the likelihood two or more events will happen
concurrently.
Problem # 1
I. What is the probability that a particular driver had exactly two speeding
violations?
II. What is the probability that a particular driver had more than two speeding
violations?
III. Which concept of probability did you use to make this estimate?
PROBLEM # 2
Three tables listed below show “random variables” and their “probabilities”. However,
only one of these is actually a probability distribution.
a. Which is it?
Tab. # 1 Tab. # 2 Tab. # 3
x P(x) x P(x) x P(x)
5 .3 5 .5 5 .2
10 .4 10 .3 10 .5
15 .2 15 -.2 15 .4
20 .1 20 .4 20 .3
b. Using the correct probability distribution, find the probability that x is:
(1) Exactly 20 (2) No more than 15 (3) More than 10 (4) At least 15
PROBLEM # 3
What is the probability that a card chosen at random from a standard deck of cards
will be either a king or a heart?Draw a A Venn diagram portrays these outcomes, which are not
mutually exclusice.
Solution
We may be inclined to add the probability of a king and the probability of a hart. But this
creates a problem. If we do that, the kign of hearts is couted with the kings and also with the
hearts. So, if we simply add the probadility of a king (there are 4 in a deck of 52 cards) to
the prbadility of a heart (there are 13 in a deck of 52 cared) and report that 17 out of 52
cards meet the requirement, we have counted the king of hearts twice. We need to subtract 1
card from the 17 so the king of hearts is counted only once. Thus, there are 16 cards that
are either hearts or kings. So the probability is 16/52=.3077.
From formula,
P(A or B) = P(A) + P(B) - P (A and B)
= 4/52 + 13/52 - 1/52
= 16/52, or .3077
A Venn diagram portrays these outcomes, which are not mutually exclusice.
Approaches to Probability
Objective Subjective
Based on available
information
Classical Empirical
Probability Probability