|
Size: 1007
Comment: Content
|
← Revision 5 as of 2025-08-06 02:13:34 ⇥
Size: 1276
Comment: Simplification
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 3: | Line 3: |
| A '''conditional probability''' is the likelihood of an event happening given that another event happens. The math notation is ''P(A|B)'', as in the probability of ''A'' given ''B''. | A '''conditional probability''' is the likelihood of an event happening given that another event happens. |
| Line 10: | Line 10: |
| == Decomposition == | |
| Line 12: | Line 11: |
| {{attachment:decomposition.svg}} | == Description == |
| Line 14: | Line 13: |
| This is the probability of an event given that some event(s) has (have) already occurred. This is generally notated as ''P(A|B)'' or ''P(A;B)'', where ''B'' has already occurred. | |
| Line 15: | Line 15: |
=== Bayes Theorem === |
It is generically decomposed as ''P(A|B) = P(A∩B) / P(B)''. Importantly though, '''Bayes theorem''' provides the following decomposition based on [[Statistics/JointProbability|joint probabilities]]: |
| Line 22: | Line 21: |
| ---- | === Independence === If two events are [[Statistics/JointProbability#Independence|independent]] (notated as ''A⫫B''), then probabilities of one do not change from being conditioned on the other. Put simply, if the conditioning probability is not 0, then: * ''P(A|B) = P(A)'' * ''P(B|A) = P(B)'' A conditioning probability of 0 will cause the conditional probability to be undefined. |
| Line 26: | Line 33: |
| == Independence == If two events are [[Statistics/Independence|independent]], then probabilities of one do not change from being conditioned on the other. Put simply, if the conditioning probability is not 0, then: ''P(A|B) = P(A)'' ''P(B|A) = P(B)'' A conditioning probability of 0 will cause the conditional probability to be undefined. ---- == Conditional Independence == |
=== Conditional Independence === |
| Line 45: | Line 36: |
| * ''P(A|B,C) = P(A|C)'' * ''P(A,B|C) = P(A|C) P(B|C)'' |
|
| Line 46: | Line 39: |
| ''P(A|B,C) = P(A|C)'' ''P(A,B|C) = P(A|C) P(B|C)'' This interrelation is sometimes notated as ''((A⫫B)|C)''. |
This interrelation is sometimes notated as ''(A⫫B)|C''. |
Conditional Probability
A conditional probability is the likelihood of an event happening given that another event happens.
Description
This is the probability of an event given that some event(s) has (have) already occurred. This is generally notated as P(A|B) or P(A;B), where B has already occurred.
It is generically decomposed as P(A|B) = P(A∩B) / P(B). Importantly though, Bayes theorem provides the following decomposition based on joint probabilities:
Independence
If two events are independent (notated as A⫫B), then probabilities of one do not change from being conditioned on the other.
Put simply, if the conditioning probability is not 0, then:
P(A|B) = P(A)
P(B|A) = P(B)
A conditioning probability of 0 will cause the conditional probability to be undefined.
Conditional Independence
If events A and B are conditionally independent, then:
P(A|B,C) = P(A|C)
P(A,B|C) = P(A|C) P(B|C)
This interrelation is sometimes notated as (A⫫B)|C.