In: Statistics and Probability
An Izod impact test was performed on 20 specimens of PVC pipe. The obtained standard deviation was s = 0.328. Assume that the random sample was drawn form a normal population N (µ, σ^2 ), with unknown mean value µ and unknown variance σ^2 .
a) Give a 99% confidence interval for the variance σ^2 .
b) Give a 99% upper confidence bound for the variance σ^2 .
c) Test the hypothesis
H0 : σ^2 = 0.1,
against the two-sided alternative
H1 : σ^2 ≠ 0.1,
at the significance level α = 0.01.
d) Explain the connection between your two-sided confidence interval and the significance test.
a) Give a 99% confidence interval for the variance σ^2 .
| 99% CI for σ using Chi-Square  | 
| (0.230, 0.547) | 
b) Give a 99% upper confidence bound for the variance σ^2 .
Upper confidence bound = 0.547
c) Test the hypothesis
H0 : σ^2 = 0.1,
against the two-sided alternative
H1 : σ^2 ≠ 0.1,
at the significance level α = 0.01.
The hypothesis being tested is:
| Null hypothesis | H₀: σ² = 0.1 | 
| Alternative hypothesis | H₁ : σ² ≠ 0.1 | 
| Method | Test Statistic  | 
DF | P-Value | 
| Chi-Square | 20.44 | 19 | 0.737 | 
The p-value is 0.737.
Since the p-value (0.737) is greater than the significance level (0.01), we cannot reject the null hypothesis.
Therefore, we have insufficient evidence to conclude that σ² = 0.1.
d) Explain the connection between your two-sided confidence interval and the significance test.
The confidence interval does not contain 0.1 and from the hypothesis test, we have insufficient evidence to conclude that σ² = 0.1. Therefore, both methods gives the same answer.