0
$\begingroup$

I'm trying to figure out the method required to solve this problem, so I stripped out the actual values to keep from getting a direct answer.

A random sample of $n$ people who walk to work are chosen, what is the probability that at least $r$ of them are injured, given that the probability of being injured while walking to work is $p$.

I don't know where to go. It feels like a binomial probability problem, but it covers a range of trials and not just one value exactly. My guess was to calculate 1-BinomCDF(n, p, r - 1). Does this seem accurate? For example, if $n = 15$, $r = 7$, $p = 0.5$.

I would have 1-BinomCDF(15, 0.5, 6) or $$1-\sum_{i=1}^6 {15\choose i}0.5^i(1-0.5)^{15-i}.$$

$\endgroup$
3
$\begingroup$

You are right, this does use the binomial distribution.$$Pr(X=r)={n\choose r}p^r(1-p)^{n-r}$$ That simply gives the probability that "r" events are true from a total of "n" possible events, with the probability of the event happening being "p"

So using your values:$$\sum_{i=7}^n Pr(X=i)$$ $$=\sum_{i=7}^n{15\choose i}0.5^i(1-0.5)^{15-i}$$

$\endgroup$
7
  • $\begingroup$ $Pr(X=7)={15\choose 7}0.5^7(1-0.5)^{15-7}$ just gives the probability for the amount injured = 7, but not the range of 7 to n, correct? $\endgroup$ – joe_04_04 Feb 9 '18 at 22:14
  • $\begingroup$ My bad, you are correct. So what you need to do is add up the all the probabilities above and including r=7 $\endgroup$ – Dylan Zammit Feb 9 '18 at 22:14
  • $\begingroup$ $1-\sum_{i=1}^6 {15\choose i}0.5^i(1-0.5)^{15-i}$? $\endgroup$ – joe_04_04 Feb 9 '18 at 22:19
  • 1
    $\begingroup$ yes, that is correct $\endgroup$ – Dylan Zammit Feb 9 '18 at 22:20
  • 1
    $\begingroup$ $1-\sum_{i=1}^6 {15\choose i}0.5^i(1-0.5)^{15-i}$ and $=\sum_{i=7}^n{15\choose i}0.5^i(1-0.5)^{15-i}$ are the same thing $\endgroup$ – Dylan Zammit Feb 9 '18 at 22:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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