I have a problem I'm not sure how to solve. It goes something like:

Suppose we are manufacturing motherboards and we take a sample measuring their
areas: [400, 400.1, 400.2, 399.99, 399.1, 400.3, 399.2, 399.99]. Give a 99% 
tolerance interval that will contain 95% of all manufactured cards (suppose it 
is a normal distribution).

From the information Provided:

n = 8
X (Sample mean) = 399.86 
S (Sample STDEV) = 0.4525

Now, I know how to get 99% confidence interval for the mean using a small sample with the T distribution but I think that includes the whole population.

Question: How can I solve this problem?

I've found examples here and there but they are for Bernoulli distributions (success vs failure proportions).
The only relevant example I found is this, but I'm not sure how to use the given formulas to solve my problem.


This article/book has all the information required to solve the problem.
The tolerance interval can be found with:

Lower: X-KS
Upper: X+KS

Using the K table provided in the article:

K(n=8, alpha=0.05, gamma=0.99) = 4.287
Lower side: 399.86 - 4.287*0.4525 = 397.9201
Upper side = 399.86 + 4.287*0.4525 = 401.7999

Where alpha is obtained from:

100(1-alpha) = 95% of the motherboards
1-alpha = 0.95
alpha = 0.05

And gamma is the confidence level.


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.