Conditional Characteristic Function

2.2k Views Asked by At

Given I know the joint characteristic function of the random variables $X,Y$ and the characteristic function of $Y$, is there a way to recover the characteristic function of $X|Y$ without inverting the characteristic functions?

Thanks...

1

There are 1 best solutions below

1
On

The joint characteristic function of $(X,Y)$ encodes the characteristic function of $Y$ hence, given the former, the latter is useless. When $Y$ is discrete, $$ E[\mathrm e^{\mathrm ixX}\mid Y]=\frac{a_x(Y)}{a_0(Y)}\qquad\text{with}\qquad a_x(y)=\int_0^{2\pi}E[\mathrm e^{\mathrm i(xX+tY)}]\,\mathrm e^{-\mathrm ity}\,\mathrm dt. $$