Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

A random variable $X$ has a distribution given by $$P(X=i)=\pi_i$$ For $i=1,2,...,p$ where $\sum_{i=1}^p n_i=1$. In a sample size $n$ the frequency of outcome $i$ is $n_i$ where $n_i\geq 0$ and $\sum_{i=1}^p n_i=n$. How do you find the maximum likelihood estimators of the $\pi_i$?

So so far I have tried this: $$L(\pi_1,...,\pi_p;x_1,...,x_n)=\prod_{i=1}^p \prod_{k=1}^{n_i}P(X_k=i)=\prod_{i=1}^p \prod_{k=1}^{n_i}\pi_i=\prod_{i=1}^p \pi_i^{n_i}$$

$$\Rightarrow l(\pi_1,...,\pi_p;x_1,...,x_n)= \sum_{i=1}^p n_i \operatorname{log}\pi_i$$ Then I say that we need for $j=1,...,p$ $$\frac{\partial l}{\partial \pi_j}=\frac{n_j}{\pi_j}+\sum_{i=1}_{i\neq j}^p \operatorname{log}\pi_i=0$$

Here I get stuck. Any help would be greatly appreciated

share|cite|improve this question
$\partial l/\partial\pi_j$ is not what you write. – Did Feb 21 '12 at 7:09
up vote 2 down vote accepted

If these are independent draws then you simply have a multinomial. The likelihood estimator is simply $\hat{\pi_i}=\frac{n_i}{n}$

share|cite|improve this answer
Thanks so much! – Freeman Feb 21 '12 at 19:19

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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