3
$\begingroup$

I have come across a question in my notes and I do not understand how to solve it. I have looked at the solutions and I still am unclear!

This is the question: enter image description here

And this is the solution:

enter image description here

I really don't understand what the question is really asking - would anyone be able to explain to me with an example? I also don't understand what the implication of the conditional probability is: e.g. the prob of 1 goal being scored given that 1 or more goals have already been scored? But then that should be independent?

Sorry - again, I am very confused and any help would be much appreciated!

Improve this question
$\endgroup$
1
  • $\begingroup$ It is more like the probability of exactly 1 goal is scored in total given that 1 or more goals in total are scored so the expected proportion of those games where goals are scored which have exactly one goal $\endgroup$
    – Henry
    Jan 24 at 13:52

2 Answers 2

Reset to default
5
$\begingroup$

The question has asked, given $N\sim\textrm{Poi}(\lambda), $ to find the value of $\lambda$ such that $\mathbb P[N=1|N\geq 1]= 0.4.$

The solution is nothing but applying the definition of conditional probability and Poisson distribution:

\begin{align}\mathbb P[N=1|N\geq 1]&=\frac{\mathbb P[(N=1)\cap(N\geq 1)]}{\mathbb P[N\geq 1]}\\&= \frac{\mathbb P[N=1]}{\mathbb P[N\geq 1]}\\&= \frac{\frac{e^{-\lambda}\lambda^1}{1!}}{1-\mathbb P[N=0]}\\&=\frac{\lambda e^{-\lambda}}{1-e^{-\lambda}}.\tag 1\label 1\end{align} Equate $\eqref 1$ with $0.4$ and find the value of $\lambda.$

Improve this answer
$\endgroup$
4
  • $\begingroup$ Thanks for your answer! Do you mind explaining why P[(N=1) n (N>=1)] = P[N=1]? $\endgroup$
    – Anna
    Jan 24 at 13:58
  • $\begingroup$ See, Anna, it is the probability of intersection of two events, one being that $N$ is equal to $1$ and other being that $N$ greater than or equal to $1.$ What should be the common event, you think? $\endgroup$
    – User1865345
    Jan 24 at 14:05
  • 3
    $\begingroup$ (+1) for the clarity! I wonder if it would be useful to also say something about how to solve (1) in $\lambda$... (the OP seem to be confused about it as well ?) $\endgroup$
    – utobi
    Jan 27 at 18:06
  • 1
    $\begingroup$ I could have @utobi but OP didn't respond further. :-) $\endgroup$
    – User1865345
    Jan 27 at 22:21
0
$\begingroup$

Conditional probability is a measure of a set (N==1) occupies upon the (N>=1) partition (rather than the whole unity, as it is for unconditional probability). Understanding this fact makes conditional probability much more intuitive.

In a physical sense, among all the cases one or more events of interest happened within a given time interval, 40% of the cases only contained one event. Now, the question is, what was the average event rate per time interval (lambda)?

Next, Poisson is a discrete distribution, so we can calculate the probability of N being 1 using Poisson PMF (lambda ^ 1 * exp(-lambda) / 1!) Probability of N>=1 is also easy to be calculated as 1 - PMF(0).

Calculating via 'trial and error' is of course not the best exercise they could think of but at least it's doable.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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