0
$\begingroup$

I have the density of a mixed r.v $X$. $$f(x)=\begin{cases} x/2 & 0\leq x <1 \\ 1/4 & x=1 \\1/4 & x=3 \\ 0 & \text{otherwise}, \end{cases}$$

How do I and calculate $P(X=2)$ and $P(X=3)$? I was thinking that $P(X=2)=F_X(2)-F_X(1)$? If this is the case, should I include the probability of the continuos part in $0\leq x <1 $ by integrating or what?

$\endgroup$
  • 1
    $\begingroup$ By integration. You just need to expand to allow for distributions. This is also exactly the same as another question just some week ago which makes me suspect it is a homework question. $\endgroup$ – mathreadler Sep 14 '17 at 18:44
  • $\begingroup$ Like which one? $\endgroup$ – Biggiez Sep 14 '17 at 18:53
  • $\begingroup$ en.wikipedia.org/wiki/Distribution_(mathematics) $\endgroup$ – mathreadler Sep 14 '17 at 19:36
0
$\begingroup$

Your $f$ is not a PDF, since $\int_{-\infty}^\infty f(x)\>dx={1\over4}\ne1$. Maybe you had the intention that the random variable $X$ in addition has "point masses" ${1\over4}$ at $x=1$ and at $x=2$. But this would still make the total mass only ${3\over4}$.

In any case such a "mixed" random variable does not have a Lebesgue integrable PDF $f$.

$\endgroup$
  • $\begingroup$ Ok is this how I calculate $P(X=2)=F_X(2)-F_X(1)=( 1/4 + \int_0^1 x/2) - (1/4 + \int_0^1 x/2 )=0$ $\endgroup$ – Biggiez Sep 14 '17 at 18:52

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.