Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Suppose that we have a network like the one in figure BN

in which we know all the conditional probabilities $P(A_i | B_j)$.

Is it possible to compute the two conditional probability $P(B1 | B2)$ and $P(B2 | B1)$?

How?

Thank you in advance.

share|improve this question
add comment

1 Answer

up vote 1 down vote accepted

From the definition of conditional probabilities: $$ P(B_2|B_1) = \frac{P(B_2 \cap B_1)}{P(B_1)} $$ Given whether $A_1$ and $A_2$ are true, $B_1$ and $B_2$ become independant events, so that we can write $$ = \frac{\sum_c P(B_2|c)P(B_1|c)}{\sum_c P(B_1|c)} $$ where the sum is over all possible combinations of $(A_1,A_2)$, $(A_1,!A_2)$...

As all $P(B_i|(!)A_1,(!)A_2)$ are known, this can be calculated.

share|improve this answer
1  
sorry that this is not more explanatory. I'm not that good with this stuff. Maybe someone else can clarify. –  example Apr 10 '12 at 16:54
    
Thank you for your help. Just a question; i can't understand what you mean with $c$ and $\sum_{c}$. What event $c$ is? –  Aslan986 Apr 10 '12 at 17:09
    
$c\in\{(A_1,A_2),(A_1,!A_2),(!A_1,A_2),(!A_1,!A_2)\}$. Did not want to write out all summands (even though there are only 3 as $!A_1,!A_2$ implies $!B_1$ and $!B_2$?). So eg the denominator expands to $(...)/(P(B_1|A_1,A_2)+P(B_1|A_1,!A_2)+P(B_1|!A_1,A_2)+P(B_1|!A_1,!A_2))$ –  example Apr 10 '12 at 17:12
    
Ok, got it now, thank you again. –  Aslan986 Apr 10 '12 at 20:46
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.