Assume we have some conditional pdf given by: $$p(x,y\mid z,c)$$


$$p(x,y\mid z,c)=p(x\mid y,z,c)p(y\mid z,c)$$

What is th reasoning behind this formula? Could you give me step by step explanation? What is exactly joint probability density of $x,y$ given some $z,c$ (what properties do we use to derive above formula)?

Second question (similiar in some way): $$p(x\mid z,c)=\int_\theta p(x,\theta\mid z,c) \, d\theta$$

What properties exactly do we use here? Obviously I have intuition behind these formulas (marginal and conditional probability rules), but I want exact properties which were used in derivation above.


Ok so I have answer for first question:

Assume that x,y=A and z,c=B, then $$p(x,y|z,c)=p(A|B)=\frac{p(A,B)}{p(B)}=\frac{p(x,y,z,c)}{p(z,c)}=\frac{p(x|y,z,c)p(y|z,c)p(z|c)p(c)}{p(z|c)p(c)}=p(x|y,z,c)p(y|z,c)$$ (by joint probability rule and chain rule of probability)

And I solved second question in similiar manner: $$p(x|z,c)=p(x|A)=\frac{p(x,A)}{p(A)}=\frac{p(x,z,c)}{p(z,c)}=\frac{\int{p(x,\theta,z,c)d\theta}}{p(z,c)}=\frac{\int{p(D,F)d\theta}}{p(z,c)}=\frac{\int{p(x,\theta|z,c)p(z,c)d\theta}}{p(z,c)}=\frac{p(z,c)\int{p(x,\theta|z,c)d\theta}}{p(z,c)}=\int{p(x,\theta|z,c)d\theta}$$

where $D=x,\theta$

| cite | improve this answer | |

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.