Engineering Data Analysis

Notes # 9

Total Probability Law

Suppose we have a partition A and A’.

Then any other event B is a union of its pieces. A B A’
That is,

B = ( A  B )  ( A'B ) A∩B A’∩B

The pieces are disjoint, so that

P(B ) = P( A  B )  ( A'B )

P(B ) = P( A  B ) + P( A'B )

By the multiplicative rule of probability,

P(B ) = P( A)P(B A) + P( A')P(B A')

In general, if the events A1 , A2 ,, An constitute a partition of the sample space S such that
P(B i )  0 for i = 1, 2, . . . , n, then for any event B of S,

P(B ) = P( A1 )P(B A1 ) + P( A2 )P(B A2 ) +  + P( An )P(B An )

Example 1: In a certain region of the country it is known from past experience that the
probability of selecting an adult over 40 years of age with cancer is 0.05. If the probability of a
doctor correctly diagnosing a person with cancer as having the disease is 0.78 and the probability
of incorrectly diagnosing a person without cancer as having the disease is 0.06, what is the
probability that an adult over 40 years of age is diagnosed as having cancer?
Solution:

Example 2: Three machines X, Y, and Z produce respectively 50%, 30%, and 20% of the total
number of items of a factory. From past data, the percentages of defective items coming from
machines X, Y, and Z are 3%, 4%, and 5% respectively. If an item is selected at random from the
factory, what is the probability that the item is defective?
Solution:
Bayes’ Theorem

Bayes' theorem, named after the English mathematician Thomas Bayes (1701 –1761), shows the
relation between two conditional probabilities which are the reverse of each other.

By the definition of conditional probability of A given that B occurred, we have

P( A  B )
P( A B ) =
P(B )
Using the total probability law, we get

P( A)P(B A)
P( A B ) =
P( A)P(B A) + P( A')P(B A')
In general, for an event A j conditioned on B ,

(
P Aj B =) P(AP(B) B)
j

Using the total probability law, we get

P(A j )P(B A j )
(
P Aj B =) P( A1 )P(B A2 ) + P( A2 )P(B A2 ) +  + P( An )P(B An )

Example 3: A rare disease exists with which only 1 in 500 is affected. A test for the disease
exists, but of course it is not infallible. A correct positive result (patient actually has the disease)
occurs 95% of the time, while a false positive result (patient does not have the disease) occurs 1%
of the time. If a randomly selected individual is tested and the result is positive, what is the
probability that the individual has the disease?
Solution:
Example 4: A truth serum has the property that 90% of the guilty suspects are properly judged
while, of course, 10% of the guilty suspects are improperly found innocent. On the other hand,
innocent suspects are misjudged 1% of the time. If the suspect was selected from a group of
suspects of which only 5% have ever
committed a crime, and the serum indicates that he is guilty, what is the probability that he is
innocent?
Solution:

Example 5: A construction company employs two sales engineers. Engineer 1 does the work of
estimating cost for 70% of jobs bid by the company. Engineer 2 does the work for 30% of jobs
bid by the company. It is known that the error rate for engineer 1 is such that 0.02 is the
probability of an error when he does the
work, whereas the probability of an error in the work of engineer 2 is 0.04. Suppose a bid arrives
and a serious error occurs in estimating cost. Which engineer would you guess did the work?
Explain.
Solution: