0
$\begingroup$

Claim: $60\%$ of the light bulb will last more than $950$ hours.

To test the claim, $100$ bulbs were randomly tested and $68$ of them lasted more than $950$ hours. Can the claim be worthy vs $0.05$ level of significant?

I can't seems to get the $z$ value as I can't get the sample deviation. So how do we approach this problem ?

$\endgroup$
0
$\begingroup$

For one bulb, either it lasts longer than $950$ hours ("success"), or it doesn't ("failure"), i.e. for one bulb we have a Bernoulli random variable. The premise of the question is that the this is a Bernoulli RV with $p=0.6$.

If we draw multiple draws from a Bernoulli and count how many come out as "success" we have a RV with Binomial distribution. In this case we would have $B(100,0.6)$ since we are testing 100 bulbs and the assumed success probability is $0.6$. If you know a hypothesis test for Biomial RVs you're finished, otherwise we can take use a test for a normal RV.

For large sample sizes the binomial distribution $B(n,p)$ can be approximated by a normal distribution $N(np,\, np(1-p))$. Thus we can apply a hypothesis test for normal RVs to the normal distribution $N(60, \,24)$ at a confidence interval of $0.95$.

$\endgroup$

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.