CONDITIONAL PROBABILITY

For an event B given that A will or has occurred, the conditional probability is given as:

𝐵 𝑃(𝐵∩𝐴)
P( ) =
𝐴 𝑃(𝐴)

Example:

The Dean of the College of Engineering at Technological University of the Philippines feels
that students who come to college with a specific goal do better than those who have no goal.
To test this theory, the Dean conducted a survey of 250 graduating students from different
courses and found out such results:

Students with Will Graduate Will not graduate

Goal 140 35

No goal 33 42

Let : A represents the event “ will graduate “

B represents the event “ students with a goal “

Suppose we want to find the probability that a student who entered college with a goal will
graduate. From the given table, of the 175 students who entered college with a goal, there are
140 who did graduate. From the desired probability is 140 / 175 = 0.8

Note that this is larger number than the probability that a student “ will graduate “ as

173 / 250 = 0.692, because we had additional information which reduced the sample space. In
other words, we found the probability that a student would graduate given that he/she entered
college with a goal. This is called the conditional probability of event A given that event B has or
will occurs, written:

𝑃(𝐴 ∩ 𝐵) 140/250 140

P ( A/B ) = = = = 0.8
𝑃(𝐵) 175/250 175
BAYE’S THEOREM

𝑃 ( 𝐴 ∩𝐵 )
From Conditional Probability, P ( A/B ) = if we multiply on both sides of the
𝑃(𝐵)
equation by P( B ), we get the formula; P( A ∩ 𝐵 ) = P( B ) P( A/B ) which enables us to
calculate the probability that two events will both occur. Similarly; P( A ∩ 𝐵 ) = P( A ) P( B/A ).
In words, the probability that two events will both occur is the product of the probability
that one of the events will occur and the conditional probability that the other event will occur
given that the first event has occurred ( occurs or will occur ). As it does not matter which
event is referred to as A and which is referred as B.

Since there are many problems in statistics which involve such pair of conditional
probabilities, let us find a formula which expresses P( A/B ) in terms of P( B/A ) for any two
events A and B. Equating the two expressions for P( A ∩ B ). We have:

P( A ) P( B/A ) = P( B ) P( A/B )

and hence;
𝐴
𝑃( 𝐵 )𝑃(𝐵)
P( B/A ) =
𝑃( 𝐴 )

If B1, B 2, B3………..and B K are mutually exclusive events of which one must occur, then:

𝐴
𝑃( 𝐵𝑖 ) 𝑃(𝐵𝑖)
P( Bi/A ) = 𝐴 𝐴 𝐴
𝑃( 𝐵1 )𝑃(𝐵1) + 𝑃( 𝐵2 ) 𝑃(𝐵2)+ ……………….. + 𝑃( 𝐵𝑘 ) 𝑃(𝐵𝑘)

for i = 1, 2……….k

And this formula is known to be as the Baye’s Rule.

MATHEMATICAL EXPECTATIONS AND DECISIONS

If the probabilities of obtaining the amounts a 1, a2, a3 ,….………….. and a k are

p1, p2, p3, ……….and pk, then the mathematical expectation is:

E = a1p1 + a2p2 + a3p3 + ………. + akpk

E = ∑ 𝑎𝑝

Note: a’s are positive when they represent profits, winnings or gains or this are the amounts
that we receive; a’s are negative when they represent losses, penalties or deficits or
this are amounts which we have to pay.

DECISION MAKING

If we have to choose between two or more alternatives, it is considered “ rational “ to

select the one with the “ most promising “ mathematical expectation: the one which maximizes
expected profits and minimizes expected costs, maximizes tax advantages and minimizes
expected losses and so on.

Example:

A furniture manufacturer must decide whether to expand his plant capacity now or wait at
least another year. His advisors tell him that if he expands now and economic conditions remain
good, there will be a profit of P 3,280,000 during the next fiscal year; if he expands now and
there is a recession there will be a loss of P 800,000. If he waits at least another year and
economic conditions remain good, there will be a profit of P 1,680,000 and if he waits a least
another year and there is a recession, there will be a small profit of P 160,000. If the furniture
manufacturer feels the probabilities for economic conditions remaining good or there being a
recession are 1/3 and 2/3 respectively, would expanding his plant capacity now maximized his
expected profit?

Suppose that an economic expert feels that the probabilities for economic conditions
remaining good be 0.4 and there being a recession be 0.6, what advised will maximize the
furniture manufacturer’s expected profit?

Suppose that the furniture manufacturer is told by his accountant that the P 3,280,000
figure is incorrect and that it should be P 3,520,000 and based on the economic expert’s
appraisal of the probabilities, will this affect his decisions?
Solution:

a) Expand Now:

E = P 3,280,000 ( 1/3 ) - P 800,000 ( 2/3 ) = P 560,000

Expand one year later:

E = P 1,680,000 (1/3) + P 160,000 ( 2/3 ) = P 666,666.67

Therefore: NO….. one year later will give the maximum profit

b) Expand Now:

E = P 3,280,000 ( 0.4 ) - P 800,000 ( 0.6 ) = P 832,000

Expand one year later:

E = P 1,680,000 ( 0.4 ) + P 160,000 ( 0.6 ) = P 768,000

Therefore; Expand now.