0
$\begingroup$

By the bayes theorem is true the following statement?

$P(S | X ,U) \propto P(X | S, U)P(S|U)$

Thanks

$\endgroup$
2
$\begingroup$

Well, it depends on what you mean by "proportional to". In the trivial sense, yes, you can divide one expression by the other, and for any given distribution you'll get some number. The constant will not be independent of the probability distribution, though.

We can explicitly calculate it with a conditional variant of Bayes rule, with $X$ and $Z$ conditional on $Y$, or equivalently, expanding the expression $P(X, Z | Y)$ in two different ways:

$$P(X | Z,Y) P(Z | Y) = P(X, Z | Y) = P(Z | X, Y) P(X | Y)$$

So $P(Z | Y)$ (or its inverse) is the constant of proportionality, but this is different for different distributions.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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