0
$\begingroup$

Suppose that we have a sequence of observations $y_1, y_2, ... y_n$ that we know is generated by $ax_1, ax_2, ...ax_n$ where $X \sim \mathrm{Poisson}$. Can we use MLE to estimate $a$ and the $\lambda$ of the Poisson distribution?

EDIT: I deleted my attempt at the MLE, it was not helpful.

$\endgroup$
1
$\begingroup$

Your likelihood is not correct, so everything you've computed makes no sense.

If $Y_i = aX_i$ where each $X_i \sim \operatorname{Poisson}(\lambda)$, then $\Pr[Y_i = y] = \Pr[X_i = y/a]$, with the obvious ramification that if $y/a$ is not a nonnegative integer, the probability is zero. Thus the joint likelihood is $$\mathcal L(a, \lambda \mid \boldsymbol y) = \prod_{i=1}^n e^{-\lambda} \frac{\lambda^{y_i/a}}{(y_i/a)!} \mathbb 1 (y_i/a \in \mathbb Z^+ \cup \{0\}).$$

From a practical perspective, if $a$ could be any real number, given a sufficiently large $n$ we can exploit the integer support of $X$ to estimate $a$ with great precision. If $a$ is itself an integer, we simply look for the least common divisor of the observations.

$\endgroup$
  • $\begingroup$ Aha, I see it, thank you. Supposing that $a$ can indeed be any real number, do you mind going through how to estimate $a$? $\endgroup$ – Ben S. Feb 22 '18 at 1:59
  • $\begingroup$ Think about it. If I gave you the sample $$\{66.0599, 28.3114, 0., 9.43712, 28.3114, 28.3114, 9.43712, 18.8742, 28.3114, 18.8742\}$$ and told you it came from such a distribution, what would your estimate of $a$ be? Wouldn't your first step be to figure out the pairwise ratios? $\endgroup$ – heropup Feb 22 '18 at 2:03
  • $\begingroup$ Yes. I guess I should fess up that I have a more complicated estimation of parameters in mind than how I constructed the question (each observation is, instead of $aX$, made from $aX + bY + cZ$ where all of $X$, $Y$, and $Z$ are Poisson. I was hoping understanding how to get $a$ by MLE in the simpler case would let me see how to solve the more complicated question. $\endgroup$ – Ben S. Feb 22 '18 at 2:10
  • $\begingroup$ But maybe I should just pose the fuller question as a question? What do you think? $\endgroup$ – Ben S. Feb 22 '18 at 2:12

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.