Conditional probability $f(Y|X=x)$?

88 Views Asked by At

Let X be a uniform random variable on the interval [0,1]. Conditioned on the event X =x, let Y be an exponential random variable with mean 1/x

I have a question that how to find the $ f \big(Y | X=x\big) $?

It is apparent that $f_{x}\big(x\big)=1$, $0 < x< 1 $ and $0$ otherwise

$$ Y = \frac{1}{x} * e^{ \big(\frac{-y}{x}\big)} $$

Appreciate the help, thanks in advance

1

There are 1 best solutions below

1
On BEST ANSWER

I had a question on how you would find the f(Y|X=x)?

You are told that $Y$ conditioned on $X=x$ (for any $x\in(0..1)$) is exponentially distributed with a mean of $1/x$.

Now an exponential distributed random variable with a mean of $1/\lambda$ has a probability density function of $\lambda\mathrm e^{-\lambda y}\mathbf 1_{y\geq 0}$.   Not much changes when we are talking about a conditional distribution; we merely need to note where the condition is supported too.$$\begin{align}\because\quad&(Y\mid X{=}x)~\sim~\mathcal{Exp}(x)\tag{Where $0<x<1$}\\[2ex]\therefore\quad& f_{\small Y\mid X=x}(y)=x\mathrm e^{-xy}\mathbf 1_{0\leqslant y\,,\, 0<x<1}\end{align}$$


-I know that fx(x)=1, 0 < x< 1 and 0 otherwise

Yes.   So you may next find the marginal pdf for $Y$.

$$\begin{align}f_{\small Y}(y)&=\int_\Bbb R f_{\small X}(x)~f_{\small Y\mid X{=}x}(y)~\mathrm d x\\[1ex]&=\mathbf 1_{0\leqslant y}~\int_0^1 x~\mathrm e^{-xy}~\mathrm d x\end{align}$$