Conditional Probability Flashcards
Definition
P(A|B) = P(A intersection B)/P(B)
P(S|A)
P(S|A) = 1
True or False
P(B|A^c) = 1 - P(B|A)
False
P(B^c|A) = 1 - P(B|A)
True or False
P(B1 union B2| A) = P(B1|A) + P(B2|A) - P(B1 inter B2| A)
True
Write the conditional backwards for
P(An | A1 inter A2 inter… inter A(n-1))
P(An | A1 inter A2 inter… inter A(n-1))
*
P(A(n-1) | A1 inter A2 inter… inter A(n-2))
* … *
P(A1)
Law of Total Probability
Let B1,B2,…,Bk form a partition of S.
P(A) = SUM over 1<= i <=k P(A|Bi)P(Bi)
Bayes’ Theorem
P(A|B) = P(B|A) P(A)/P(B)
When do we use Bayes’ Theorem?
We use it when finding the other conditional is easier.
In the world of medical diagnosis P(Pos|Disease) is called…
sensititvity
n the world of medical diagnosis P(Neg|Disease) is called…
Specificity
n the world of medical diagnosis P(Disease) is called…
disease prevalence
n the world of medical diagnosis P(Disease|Pos) is called…
Positive predicted value
Notice: to find the positive predicted value of a test we need to have all three:
- P(Pos|D) sensitivity
- P(Neg|D) specificity
- P(D) disease prevalence