Ask your own question, for FREE!
Statistics 24 Online
OpenStudy (anonymous):

How to get \[\frac{\lambda ^k}{k!}e^{-\lambda}\] from \[\frac{n!}{(n-k)!k!}p^k(1-p)^{n-k}\]when \(n \rightarrow \infty\)

OpenStudy (anonymous):

taking a limit i believe

OpenStudy (anonymous):

i am surprised this is not in your text, can look it up and give you a clue if you like

OpenStudy (anonymous):

first off we need to state exactly what the assumption is, namely that as \(n\to \infty\) and \(p\to 0\) you have \(np\to \lambda>0\)

OpenStudy (anonymous):

then you write \(np =\lambda+\epsilon_n\) and now the assumption is \(\epsilon_n\to 0\) ans \(n\to \infty\)

OpenStudy (anonymous):

I was trying to work it out using Stirling formula, but looks complicated (I haven't finished it yet...) Attached is what I've got in the notes, but don't quite understand it.

OpenStudy (anonymous):

yeah it is a bunch of algebra, i don't think sterling is involved

OpenStudy (anonymous):

Yes, Stirling's doesn't work..

OpenStudy (anonymous):

you can think of it this way \[\binom{n}{k}p^k(1-p)^{n-k}\] now stick the multiply and divide by the \(n^k\) and get \[\frac{(1-\frac{1}{n})(1-\frac{2}{n})(1-\frac{3}{n})...(1-\frac{(m-1)}{n})}{k!}(mp)^k\left(1-\frac{\lambda+\epsilon_n}{n}\right)^{n-k}\]

OpenStudy (anonymous):

wow i screwed that all up

OpenStudy (anonymous):

i meant pull out an \(n^k\) and stick it in the second part \[\frac{(1-\frac{1}{n})(1-\frac{2}{n})(1-\frac{3}{n})...(1-\frac{(k-1)}{n})}{k!}(np)^k\left(1-\frac{\lambda+\epsilon_n}{n}\right)^{n-k}\]

OpenStudy (anonymous):

now let \(n\to \infty\) and you get \[\frac{\lambda^k}{k!}\lim_{n\to \infty}\left(\frac{1-\lambda +\epsilon_n}{n}\right)^{n-m}\]

OpenStudy (anonymous):

i meant k, but in any case it is fixed, so the second term gives \(e^{-\lambda}\) whichever version, it is really algebraic manipulation that gets it

OpenStudy (anonymous):

Poisson is a limiting case of binomial considering infinite number of trials and constant expectation of success no matter how many trials. \[\begin{array}{l}\mu=np=cons\tan t\rightarrow p=\frac\mu n\\\lim_{n\rightarrow\infty}\left(\begin{array}{c}n\\k\end{array}\right)p^k\left(1-p\right)^{n-k}=\lim_{n\rightarrow\infty}\left(\begin{array}{c}n\\k\end{array}\right)\left(\frac\mu n\right)^k\left(1-\frac\mu n\right)^{n-k}=\\=\lim_{n\rightarrow\infty}\frac{n\left(n-1\right)...\left(n-k+1\right)}{n^k}\left(\frac{\mu^k}{k!}\right)\left(1-\frac\mu n\right)^{-k}\left[\left(1-\frac1{\displaystyle\frac n\mu}\right)^{-\frac n\mu}\right]^{-\mu}\end{array}\]where\[\begin{array}{l}\lim_{n\rightarrow\infty}\frac{n\left(n-1\right)...\left(n-k+1\right)}{n^k}=1\\\lim_{n\rightarrow\infty}\left(1-\frac\mu n\right)^{-k}=1\\\lim_{n\rightarrow\infty}\left[\left(1-\frac1{\displaystyle\frac n\mu}\right)^{-\frac n\mu}\right]^{-\mu}=e^{-\mu}\end{array}\]then,\[\lim_{n\rightarrow\infty}\left(\begin{array}{c}n\\k\end{array}\right)p^k\left(1-p\right)^{n-k}=\frac{\mu^k}{k!}e^{-\mu}\]

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!