In: Statistics and Probability
The manager of the service department of a local car dealership
has noted that the service times of a sample of 24 new automobiles
has a standard deviation of 5 minutes.
A.) Compute a 90% confidence interval estimate for the standard
deviation of the service times (in minutes) for all their new
automobiles.
B.) The manager of the service department of a local car dealership
wants to test the claim that the variance of the service time does
not exceed 27 minutes. Test this claim using a significance level
of 0.1.
Solution :
A) The 90% confidence interval for population standard deviation is given as follows :
Where, n is sample size, s is sample standard deviation and the terms given in denominator are critical values of chi-square to construct 90% confidence interval for population standard deviation.
We have, n = 24, s = 5 minutes
Using chi-square table we get,
Hence, 90% confidence interval estimate for the standard deviation of the service times (in minutes) for all their new automobiles is,
The 90% confidence interval estimate for the standard deviation of the service times (in minutes) for all their new automobiles is (4.042 minutes, 6.628 minutes).
B) Null and alternative hypotheses :
The null and alternative hypotheses would be as follows :
Test statistic :
To test the hypothesis we shall use chi-square test for variance. The test statistic is given as follows :
Where, n is sample size, s is sample standard deviation and σ² is hypothesized value of population variance under H0.
We have, n = 24, s = 5 and σ² = 27
The value of the test statistic is 21.2963.
P-value :
Since, our test is right-tailed test, therefore we shall obtain right-tailed p-value for the test statistic. The right-tailed p-value is given as follows :
P-value = P(χ² > value of the test statistic)
P-value = P(χ² > 21.2963)
P-value = 0.5630
The p-value is 0.5630.
Decision :
P-value = 0.5630
Significance level = 0.01
(0.5630 > 0.10)
Since, p-value is greater than the significance level of 0.10, therefore we shall be fail to reject the null hypothesis (H0) at 0.10 significance level.
Conclusion : At 0.10 significance level, there is not sufficient evidence to support the claim that the variance of the service time does not exceed 27 minutes.
Please rate the answer. Thank you.