0
$\begingroup$

i have a random sample of n observations from geometric distribution pdf $p q^x-1$ but i m not able to find the

1) maximum likelihood estimator of the mean ,

2) MLE of P using exponential family

as i found MLE for P but confused about MLE for mean and for P of exponential family

$\endgroup$
0
$\begingroup$

Hint:

One form of the geometric distribution has probability mass function (not density) $pq^{x-1}$ on the non-negative integers

in which case the mean is $\mu=\frac{q}{p}$ so $p = \frac{1}{\mu+1}$ and $q=\frac{\mu}{\mu+1}$ and the probability mass function is $\frac{\mu^{x-1}}{(\mu+1)^x}$

so you can find the likelihood of your $n$ observations as a function of $\mu$, $n$ and $\sum x_i$, and you can find the value of $\mu$ which maximises this.

You should also be able to find a simple relationship between your two answers.

$\endgroup$
  • $\begingroup$ plz little more elaborate, should i take likelihood log etc to solve that last quantity that u have given after putting the value of p and q $\endgroup$ – rizwan niaz Feb 11 at 23:59

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.