STA2017

PROBABILITY DISTRIBUTION
THEORY
NOTES
.
Basic Probability
Three Axioms of Probability
1.
2.
3. For events , ,,…, , if for all i≠j
Example: Theorem where is the complement of

Proof:
By axiom 1,

and by axiom 3, since and ,

then =
 Theorem For events and B,
Proof
Since and and are mutually exclusive
Then by axiom 3,

Theorem For events and B,

Proof
It can be understood that the set is made up of the union of
two mutually exclusive sets namely and . Therefore by axiom
3,
Theorem If then

Proof
If then , where and are mutually exclusive events.
By axiom 3,
But from axiom 1,
which implies that
Identifying sample Points

The elements of the sets representing the events are called

outcomes of the event or sample points.
Example
The new motor vehicle license plate system for an extremely
small community is based on issuing a limited set of licenses
with codes each of which is made up of three items; two
vowels from among {A, E, I, O, U} which may be a repeat
followed by a single digit selected from among {1, 2, 3, 4, 5,
6, 7, 8, 9, 0}. Examples of possible sample points are: UE9 or
II4 or AO0.
Counting Sample points

The general principle of counting outcomes is by

multiplying the number of options for each independent
factor variable. For example if the number of rows of seats in
a theatre is r but the number of seats in each row is n. Then
the total number of seats is rn
When counting sample points there are four categories. One
criterion is to determine the situation based on whether or not
they are with or without replacement. A second criterion is
to determine whether or not order is involved in identifying
the sample points.
Counting Sample points (continued)
Case 1 With replacement and with order
One Prize is to be awarded for each of three subjects to one
of twenty (20) students in a class. If students are not
restricted in the number of prizes that they may receive in
how many ways may the three prizes be awarded?
The number of ways to award the prizes =20 X 20 X 20 = 20 3
= 8000. In this case, n=20 r=3
 Counting Sample points (continued)
Case 2 Without replacement and with order
In the case of an athletic race in which n persons are
competing, after someone places first that individual will not
be attempting to place again. That is what is meant by
without replacement – no athlete after finishing will seek to
be replaced in the same race. Therefore, the number available
for the remaining places will reduce.
The number of ways in which the n athletes can finish is n X
(n-1) X (n-2) X …X 1 = n!
If we are only interested in the placement in r positions from
among the n athletes the number of ways would be a product
of numbers decreasing from n by 1 filling r places
= n X (n-1) X (n-2) X . . . X (n-(r-1)) = n X (n-1) X (n-2) X . . . X (n
– r +1) =
 Counting Sample points (continued)
Case 3 Without replacement and without order
Selecting a combination of r items regardless of order from
among n items
==
If you were required to select r=3 items from among n=4 items
{A, B, C, D} without replacement the number of combinations
would be
= 4 They are ABC, ABD, ACD, BCD
Counting Sample points (continued)
Case 4 With replacement without order
This is the real tricky case. Suppose we are to select 3 items
from among the 4 items ABCD but we are allowed to repeat
the items selected without considering the order in which
they are placed. Imagine that the four items are instead 4 pans
numbered A, B, C and D respectively but we are to place
these 3 items represented by markers M without restriction
within the four pans.
If we do not count the cell walls at either end then we can
count the number of arrangements of the remaining cell walls
and markers combined (n+r-1). Then in order to eliminate
redundant orderings of the items selected (the markers)
divide by the r!.
 Replacement
criteria

With Without
Ordering With
criteria

Without
Sample points :Equal and Unequal probabilities
If the events have sample points with unequal probabilities
the individual probabilities of these sample points sum to 1.
Example
A sample space S contains sample points A, B, C and D such
that P(A) = 3P(C),
P(B) = P(D) = 0.2. Find the values of each sample point.

In the case that sample points have equal probabilities then if

n(E) is the number of sample points satisfying event E and
N=n(S) is the number of sample points then
Example 1
A team consisting of one man and one woman are to be
selected from among two men and three women, with the
restriction that there is a specific man and a specific woman
within the group that are not allowed to be on the team
together.
Give a representation of this sample space
Find the probability that the team does not include the
woman who was not allowed to be in the team with the
specific man.
Solution
Let the five persons be represented by where the Ws
represent women, the Ms represent men and the subscripts
with “not” represent the individuals that are not allowed with
those represented with like subscript. The sample space
therefore is
S=
The event that the team of two does not include the woman
who was not allowed in the team with the man is given by .
The required probability is therefore 4/5=0.8
 Example 2 The three item license code described previously
as consisting of two vowels (allowed to repeat) and one of the
ten digits
What is the total number of such license codes possible?
Find the probability that a license code selected
(i) Contains no repeated vowel
(ii) Has a repeated vowel and an odd number digit
Solution
The total number of possible license codes = 5 2X 10 = 250
(iii)The number of codes with no repeated vowel
= = 20 X 10 =200
P(no repeated vowel) = = 0.8
(ii) P(repeated vowel and odd number) = = 0.1
 More popular laws of probability
If and B are mutually exclusive,
Conditional Probability

If and B are independent then

And
And
Total Probability

Bayes’ Rule

If then