In: Statistics and Probability
As part of a survey, a marketing representative asks a random sample of 3030 business owners how much they would be willing to pay for a website for their company. She finds that the sample standard deviation is $34043404. Assume the sample is taken from a normally distributed population. Construct 9090% confidence intervals for (a) the population variance sigmaσsquared2 and (b) the population standard deviation sigmaσ. Interpret the results. (a) The confidence interval for the population variance is (nothing,nothing). (Round to the nearest integer as needed.)
Solution:
Part a
Confidence interval for population variance is given as below:
(n – 1)*S2 / χ2α/2, n – 1 < σ2 < (n – 1)*S2 / χ21 -α/2, n– 1
We are given
Confidence level = 90%
Sample size = n = 30
Degrees of freedom = n – 1 = 30 – 1 = 29
Sample standard deviation = S = 3404
χ2α/2, n – 1 = 42.5570
χ21 -α/2, n– 1 = 17.7084
(By using chi square table)
(n – 1)*S2 / χ2α/2, n – 1 < σ2 < (n – 1)*S2 / χ21 -α/2, n– 1
(30 – 1)*3404^2/42.5570 < σ2 <(30 – 1)*3404^2/17.7084
7895986.9935 < σ2 < 18975734.8388
Lower limit = 7895986.9935
Upper limit = 18975734.8388
We are 90% confident that the population variance will lies within $7895987 and $18975735.
Part b
Confidence interval for population standard deviation is given as below:
Sqrt[(n – 1)*S2 / χ2α/2, n– 1 ] < σ < sqrt[(n – 1)*S2 / χ21 -α/2, n– 1 ]
Sqrt(7895986.9935) < σ < Sqrt(18975734.8388)
2809.9799 < σ < 4356.1146
Lower limit = 2809.9799
Upper limit = 4356.1146
We are 90% confident that the population standard deviation will lies within $2810 and $4356.