0
$\begingroup$

I have the joint PMF

$\exp(y_1 \ln(\lambda)+y_2 \ln(c)+y_2\ln(\lambda)-\ln(y_1!y_2!)-\lambda(1+c))$

for a constant $c>0$. In canonical representation and mixed parameterization I have $\mathbf{\theta}=(\ln(c),\ln(\lambda))$ and $\mathbf{t}=(v,u)^T$ with $v=Y_2$ and $u=Y_1+Y_2$.

What is the marginal PMF $f(v)$?

$\endgroup$
  • 1
    $\begingroup$ Do you have $y_1,y_2 \in \mathbb{N}_0$? $\endgroup$ – Dieter Kadelka Aug 6 at 10:18
  • 1
    $\begingroup$ Do you have $y_1,y_2 \in \mathbb{N}_0$? If this is true, then $Y_1$ and $Y_2$ are independent, $Y_1$ has Poisson distribution $Po(\lambda)$ and $Y_2$ has distribution $Po(\lambda \dot c)$. $\endgroup$ – Dieter Kadelka Aug 6 at 10:24
  • 1
    $\begingroup$ Yes the distributions are independent with $Y_1 \sim Po(\lambda)$ and $Y_2 \sim Po(c\lambda)$. $\endgroup$ – Orongo Aug 6 at 12:19
1
$\begingroup$

The pdf of $Po(c \dot \lambda)$ is just $v \to e^{-c\lambda} \frac{(c\lambda)^v}{v!}$, $v \in \mathbb{N}_0$.

$\endgroup$

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.