Explore BrainMass

Use Excel to construct confidence intervals

I need help in setting up and solving the problem given below. The problem with necessary data is fully set up in the attached excel document.

INSTRUCTION: Include a "summary statement" for each CI.
Use the data for which is given in columns A:C.



(A) Give a 84% CI for the mean of Brand 1 minus mean lifetime (in hours) of Brand 2, using the formula for equal population variances.

(B) Repeat, but using the general formula for any relationship between the population variances.

Use excel to do this. Excel formulas will automatically give the output for both formulas. All you have to do is figure out where the answer is for each part and clearly indicate which cell has the answer.

The book gives guidance on which of the two CI outputs should be used. The one in (B) is always correct (unless the sample sizes are small and at least one population is not normally distributed). The one in (A) is sometimes correct as well. If the number in cell is "large" then the (A) result is also correct. People differ about what "large" would mean here. Anything larger than 0.25 would satisfy about everyone. Anything larger than 0.05 would be the rule for a substantial number of analysts. A few people would use 0.01 as the lower limit for "large". We can use 0.05 for this problem.


Solution Summary

The solution uses excel to construct confidence intervals.