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.

Question 1:

From some slides I saw that:

$p(x_{0:t+1}|y_{1:t+1}) = p(x_{0:t}, x_{t+1}|y_{1:t}, y_{t+1})$, and according to the product rule of probability theory $P(a,b) = P(a|b) P(b)$ we will have have:

$$p(x_{0:t+1}|y_{1:t+1}) = p(x_{0:t}, x_{t+1}|y_{1:t}, y_{t+1}) = \frac{p(x_{0:t}, x_{t+1}, y_{t+1}|y_{1:t})}{p(y_{t+1}|y_{1:t})}$$

I don't understand how the product rule of probability theory is applied for this example, what is $a$ and $b$ in this example ?

Note: $x_{0:t}$ is $\{x_0,x_1,x_2, \ldots, x_t\}$

Question 2:

Also, is it always true that: $P(X|Y)= \int P(X|Z) P(Z|Y)\, dz$ ?

share|improve this question
add comment

1 Answer

up vote 1 down vote accepted

The product rule of probability as you state is $P(A, B) = P(A|B)P(B)$. This rule can also be applied to conditional probabilities as well i.e $P(A, B|C) = P(A|B, C)P(B|C)$. This gives, $P(A|B, C) = \frac{P(A, B|C)}{P(B|C)}$. Comparing the RHS in this equation with the equation that you have, you will see $A = x_{0:t+1}, B = y_{t+1}, C = y_{1:t}$.

As for Question 2, $P(X|Y) = \int P(X, Z|Y) dz$. Now, generally if you factorize the term in the integral, it will factorize as $P(X|Z, Y)P(Z|Y)$. However, in your expression it is factorizing as $P(X|Z)P(Z|Y)$, which implies that $X$ is conditionally independent of $Y$ given $Z$. This need not be always true, hence your expression is not always true.

share|improve this answer
    
I didn't know that the product rule can be applied this way to a conditional probability, but it's a little confusing: is $P(A,B|C) = P((A,B)|C) = P(A,(B|C))$ ? –  shn Apr 24 '12 at 18:45
    
It's always $P((A,B)|C)$. –  TenaliRaman Apr 24 '12 at 18:51
    
Ok and can we say that $P((A|B)|C) = $P(A|B,C)$ ? –  shn Apr 24 '12 at 19:40
    
'$|$' is not an operator so $P((A|B)|C)$ does not mean anything. '$|$' is simply a partition to indicate which random variables are unobserved and which ones are observed. Basically, in $P(A_1,A_2,\ldots,A_n|B_1,B_2,\ldots,B_m)$, the '$|$' indicates that $A_1,\ldots,A_n$ are unobserved random variables and $B_1,\ldots,B_m$ are the observed random variables. –  TenaliRaman Apr 24 '12 at 21:38
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.