Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I'm working through some practice exercises for my probability course and I think there's a concept I'm not totally understanding.

The question I'm trying to work through is as follows:

The number of mail a post office receives in one day is defined by a Poisson random variable with parameter $\lambda = 5$. The post office fails to process a piece of mail with probability $0.04$ independently with every ball. Find the probability that the post office processes exactly 5 pieces of mail in a working day.

From what I understand, we can model the number of calls as $\dfrac {(e^5)5^{10}} {10!}$ but how do I use the rest of the information to determine the probability mass function?

share|improve this question
2  
"The post office fails to process a piece of mail with probability 0.04 independently with every ball." What does that mean, "with every ball"? –  Imray Nov 11 '12 at 20:53
    
i think it's: "The post office fails to process a piece of mail with probability 0.04 independently with every mail received." –  Jaynathan Leung Nov 12 '12 at 1:29
add comment

2 Answers

It will turn out that the number of letters processed has Poisson distribution with parameter $\lambda(1-0.04)$. There are better conceptual ways of seeing this, but instead we do a somewhat unpleasant computation.

The Post Office receives $n$ letters with probability $e^{-\lambda}\dfrac{\lambda^n}{n!}$. Given that it receives $n$ letters, the number of letters processed has Binomial distribution. Specifically, the (conditional) probability that it processes $k$ letters is $\binom{n}{k}p^kq^{n-k}$. (In our case, $p=0.96$ and $q=1-p=0.04$.)

So the probability that the Post Office processes exactly $k$ letters is given by $$\sum_{n=k}^\infty e^{-\lambda}\frac{\lambda^n}{n!}\binom{n}{k}p^kq^{n-k}.$$ We are interested specifically in the sum for $k=5$. But we continue with the general analysis.

Use $\dbinom{n}{k}=\dfrac{n!}{k!(n-k)!}$ and some algebra to rewrite the above expression as $$e^{-\lambda}\frac{(p\lambda)^k}{k!} \sum_{n=k}^\infty\frac{1}{(n-k)!} (\lambda q)^{n-k}.$$ The $\sum_{n=k}^\infty$ part, perhaps after putting $j=n-k$, can be seen to be the power series expansion of $e^x$, with $x=\lambda q$. But $e^{-\lambda}e^{\lambda q}=e^{-\lambda p}$, so our expression simplifies to $$e^{-\lambda p}\frac{(\lambda p)^k}{k!}.$$ It follows that the number of letters processed has Poisson distribution, parameter $\lambda p$. In our case, $\lambda p=4.8$. Now finding the probability that $k=5$ is easy.

share|improve this answer
add comment

I don't know where you got $10$. You need $$\frac{e^{-\lambda} \lambda^x}{x!} = \frac{e^{-5}5^5}{5!}\text{ since }\lambda=5\text{ and } x=5.$$

share|improve this answer
add comment

Your Answer

 
discard

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.