Thread: Joint Probability

From past experience, 10 % of the people over 30 years have a certain type of disease. The probability of a doctor correctly diagnosing a person with the disease is 78% and incorrectly not having the disease 8%.

a. what is the probability that a person is diagonsed as having the disease?

b. what is the probability that a person diagonesd as having the disease actually have the disease?

Originally Posted by weakmath

From past experience, 10 % of the people over 30 years have a certain type of disease. The probability of a doctor correctly diagnosing a person with the disease is 78% and incorrectly not having the disease 8%.

a. what is the probability that a person is diagonsed as having the disease?

b. what is the probability that a person diagonesd as having the disease actually have the disease?

Let D be the event "have the disease".
Let A be the event "diagnosed with the disease".

Given data:

"10 % of the people over 30 years have a certain type of disease": Pr(D) = 0.1 => Pr(D') = 0.9

"The probability of a doctor correctly diagnosing a person with the disease is 78%": Pr(A | D) = 0.78 => Pr(A' | D) = 1 - 0.78 = 0.22

"The probability of a doctor ... incorrectly not having the disease 8%": Pr(A | D') = 0.08 => Pr(A' | D') = 0.92 (well, this is the only interpretation of the wording that makes any sense to me, anyway)


a. Pr(A) = Pr(A | D) Pr(D) + Pr(A | D') Pr(D') = ....


b. You need Pr( D | A).

Note that and .

Therefore .

This is just another way of writing the usual formula .

Now substitute the values of Pr( A | D), Pr(D) and Pr(A).