Probability of at least t time units between occurences given intensity

41 Views Asked by At

It's been 15 years ago I did studies in statistics, so most of the knowledge is gone. I've looked up Poisson's distribution but the tooth of time has taken a too large munch of my skills.

If I know that an alarm occurs 10 times a day, I have the average time between them to 144 minutes, which is my expected value. But what is the likelihood of the following events?

  1. No alarm for 144 minutes
  2. Precisely 1 alarm in 144 minutes
  3. No alarm for 288 minutes
  4. At least 3 alarms in 288 minutes

Please note that I'm not looking for the answer but rather a pointer on how to approach it. And in case this resembles school assignment and a lazy student, I'm not. However, it's a bit difficult to prove a non-fact. Perhaps a credential consisting of my diploma work with a 10+ years worth of dust on it will do.

1

There are 1 best solutions below

4
On BEST ANSWER

Are you assuming alarms are independent events with Poisson distribution? If so, it straightforward. Use $p_n=e^{-\lambda}\frac{\lambda ^n}{n!}$. where $p_n$ is the probability of $n$ alarms in an interval where $\lambda=1$ for the first two questions and $=2$ for the second two.

Note for question 4, the answer requires $\sum_{n=3}^\infty p_n$. It is easier to get $1 - \sum_{n=0}^2 p_n$.