Confidence Intervals for the Variance of a Normal Population Confidence Intervals for the Variance of a Normal Population Theorem Let X1 , X2 , . . . , Xn be a random sample from a distribution with mean µ and variance σ 2 . Then the random variable P (Xi − X )2 (n − 1)S 2 = σ2 σ2 has s chi-squared (χ2 ) probability distribution with n − 1 degrees of freedom (df). Confidence Intervals for the Variance of a Normal Population Confidence Intervals for the Variance of a Normal Population Proposition A 100(1 − α)% confidence interval for the variance σ 2 of a normal population has lower limit (n − 1)s 2 /χ2α ,n−1 2 and upper limit (n − 1)s 2 /χ21− α ,n−1 2 A confidence interval for σ has lower and upper limits that are the square roots of the corresponding limits in the interval for σ 2 . Hypotheses and Test Procedures Hypotheses and Test Procedures Definition The null hypothesis, denoted by H0 , is the claim that is initially assumed to be true (the “prior belief” claim). The alternative hypothesis, denoted by Ha , is the assertion that is contradictory to H0 . The null hypothesis will be rejected in favor of the alternative hypothesis only if sample evidence suggests that H0 is false. If the sample does not strongly contradict H0 , we will continue to believe in the truth of the null hypothesis. Hypotheses and Test Procedures Definition The null hypothesis, denoted by H0 , is the claim that is initially assumed to be true (the “prior belief” claim). The alternative hypothesis, denoted by Ha , is the assertion that is contradictory to H0 . The null hypothesis will be rejected in favor of the alternative hypothesis only if sample evidence suggests that H0 is false. If the sample does not strongly contradict H0 , we will continue to believe in the truth of the null hypothesis. Remark: There are only two possible conclusions from a hypothesis testing: reject H0 or fail to reject H0 ; or corresponding accept Ha or fail to accept Ha . We NEVER reject Ha . Hypotheses and Test Procedures Hypotheses and Test Procedures A test procedure is specified by the following: 1. A test statistic, a function of the sample data on which the decision (reject H0 or do not rejece H0 ) is to be based; 2. A rejection region, the set of all test statistic values for which H0 will be rejected. The null hypothesis will then be rejected if and only if the observed or computed test statistic value falls in the rejection region. Hypotheses and Test Procedures A test procedure is specified by the following: 1. A test statistic, a function of the sample data on which the decision (reject H0 or do not rejece H0 ) is to be based; 2. A rejection region, the set of all test statistic values for which H0 will be rejected. The null hypothesis will then be rejected if and only if the observed or computed test statistic value falls in the rejection region. Remark: The rejection region is more subjective. Hypotheses and Test Procedures Hypotheses and Test Procedures Definition A type I error consists of rejecting the null hypothesis H0 when it is true, traditionally denoted the probability of making type I error by α. A type II error involves not rejecting H0 when it is false, traditionally denoted the probability of making type I error by β. Mathematically, α = P(Type I Error) = P(reject H0 | H0 ) β = P(Type II Error) = P(fail to reject H0 | Ha ) Hypotheses and Test Procedures Definition A type I error consists of rejecting the null hypothesis H0 when it is true, traditionally denoted the probability of making type I error by α. A type II error involves not rejecting H0 when it is false, traditionally denoted the probability of making type I error by β. Mathematically, α = P(Type I Error) = P(reject H0 | H0 ) β = P(Type II Error) = P(fail to reject H0 | Ha ) Remark: There are no error-free test procedures unless we have the information for the whole population. Therefore, we look for those test for which the probability of making either type of error is small. Test about a Population Mean Test about a Population Mean Example: To determine whether the pipe welds in a nuclear power plant meet specifications, a random sample of 10 welds is selected, and tests are conducted on each weld in the sample. The sample data is recorded as follows 101.9 100.4 101.2 100.9 101.7 with X = 101.10. 101.5 100.9 100.1 101.6 100.8 It is known that the weld strength is normally distributed with mean µ and standard deviation σ = 2. If the specifications state that the mean strength should be equal to 100 lb/in2 , shall we accept that the pipe welds meet the specifications with significance level .05? Test about a Population Mean Test about a Population Mean 1. Parameter of interest: µ = population average strength. Test about a Population Mean 1. Parameter of interest: µ = population average strength. 2. Null hypothesis: H0 : µ = µ0 = 100. Test about a Population Mean 1. Parameter of interest: µ = population average strength. 2. Null hypothesis: H0 : µ = µ0 = 100. 3. Alternative hypothesis: Ha : µ 6= 100. Test about a Population Mean 1. 2. 3. 4. Parameter of interest: µ = population average strength. Null hypothesis: H0 : µ = µ0 = 100. Alternative hypothesis: Ha : µ 6= 100. Test statistic value: z= x̄ − µ0 x̄ − 100 √ = √ σ/ n 2/ n Test about a Population Mean 1. 2. 3. 4. Parameter of interest: µ = population average strength. Null hypothesis: H0 : µ = µ0 = 100. Alternative hypothesis: Ha : µ 6= 100. Test statistic value: z= x̄ − µ0 x̄ − 100 √ = √ σ/ n 2/ n 5. Rejection region: z ≥ z.025 or z ≤ −z.025 , where z.025 = 1.96. Test about a Population Mean 1. 2. 3. 4. Parameter of interest: µ = population average strength. Null hypothesis: H0 : µ = µ0 = 100. Alternative hypothesis: Ha : µ 6= 100. Test statistic value: z= x̄ − µ0 x̄ − 100 √ = √ σ/ n 2/ n 5. Rejection region: z ≥ z.025 or z ≤ −z.025 , where z.025 = 1.96. 6. Substituting n = 10 and x̄ = 101.10, z= 101.10 − 100 √ = 1.74 2/ 10 Test about a Population Mean 1. 2. 3. 4. Parameter of interest: µ = population average strength. Null hypothesis: H0 : µ = µ0 = 100. Alternative hypothesis: Ha : µ 6= 100. Test statistic value: z= x̄ − µ0 x̄ − 100 √ = √ σ/ n 2/ n 5. Rejection region: z ≥ z.025 or z ≤ −z.025 , where z.025 = 1.96. 6. Substituting n = 10 and x̄ = 101.10, z= 101.10 − 100 √ = 1.74 2/ 10 7. Since −1.96 < 1.74 < 1.96, i.e., the value of the test statistic does not fall in the rejection region (−∞, 1.96) ∪ (1.96, ∞), we can not reject H0 at significance level .05. Test about a Population Mean Test about a Population Mean Test for Population Mean of A Normal Population with Known σ Null hypothesis: Test statistic value H0 : µ = µ0 x̄−µ √0 z = σ/ n Alternative Hypothesis Ha : µ > µ0 Ha : µ < µ0 Ha : µ 6= µ0 Rejection Region for Level α Test z ≥ zα (upper-tailed test) z ≤ −zα (lower-tailed test) z ≥ zα/2 or z ≤ −zα/2 (two-tailed test) Test about a Population Mean Test about a Population Mean Example: To determine whether the pipe welds in a nuclear power plant meet specifications, a random sample of 10 welds is selected, and tests are conducted on each weld in the sample. The sample data is recorded as follows 101.9 100.4 101.2 100.9 101.7 with X = 101.10. 101.5 100.9 100.1 101.6 100.8 It is known that the weld strength is normally distributed with mean µ and standard deviation σ = 2. If the specifications state that the mean strength should be equal to 100 lb/in2 , shall we accept that the pipe welds meet the specifications with significance level .05? What is the probability of making type II error then? Test about a Population Mean Test about a Population Mean H0 : µ = µ0 v.s. Ha : µ > µ0 Then the rejection region for level α test is z ≥ zα , or equivalently √ x̄ ≥ µ0 + zα · σ/ n. Let µ0 denote a particular value of µ that is less than the null value µ0 , then β(µ0 ) = P(H0 is not rejected | µ = µ0 ) √ = P(X < µ0 + zα · σ/ n | µ = µ0 ) X − µ0 µ0 − µ 0 0 √ < zα + √ |µ=µ =P σ/ n σ/ n µ0 − µ 0 √ = Φ zα + σ/ n Test about a Population Mean Test about a Population Mean Alternative Hypothesis Ha : µ > µ0 Ha : µ < µ0 Ha : µ 6= µ0 Type II Error Probability β(µ0 ) for Level α Test 0 √ Φ zα + µσ/0 −µ n 0 µ0 −µ 1 − Φ −zα + σ/√n 0 √ Φ zα/2 + µσ/0 −µ − Φ −zα/2 + n 0 µ0 −µ √ σ/ n Test about a Population Mean Test about a Population Mean Example: To determine whether the pipe welds in a nuclear power plant meet specifications, a random sample of n welds is selected, and tests are conducted on each weld in the sample. It is known that the weld strength is normally distributed with mean µ and standard deviation σ = 2. And the specifications state that the mean strength should be equal to 100 lb/in2 . To construct a hypothesis with α = .05 and β = .1, how large should n be? Test about a Population Mean Test about a Population Mean The sample size n for which a level α test also has β(µ0 ) = β at the alternative value µ0 is h i σ(zα +z ) 2 for a one-tailed (upper or lower) test µ0 −µ0β n = σ(z +z ) 2 β α/2 for a two-tailed test (an approximate solution) µ0 −µ0 Test about a Population Mean Test about a Population Mean Example: To determine whether the pipe welds in a nuclear power plant meet specifications, a random sample of 50 welds is selected, and tests are conducted on each weld in the sample. The average strength of this sample is X = 101.10, and the standard deviation is s = 2. If the specifications state that the mean strength should exceed 100 lb/in2 , shall we accept that the pipe welds meet the specifications with significance level .05? Test about a Population Mean Test about a Population Mean Example: To determine whether the pipe welds in a nuclear power plant meet specifications, a random sample of 10 welds is selected, and tests are conducted on each weld in the sample. The sample data is recorded as follows 101.9 100.4 101.2 100.9 101.7 with X = 101.10 and 101.5 100.9 100.1 101.6 100.8 s = .585. It is known that the weld strength is normally distributed with mean µ. If the specifications state that the mean strength should exceed 100 lb/in2 , shall we accept that the pipe welds meet the specifications with significance level .05? Test about a Population Mean Test about a Population Mean Test for Population Mean of A Normal Population with Unknown σ Null hypothesis: Test statistic value H0 : µ = µ0 √0 t = x̄−µ s/ n Alternative Hypothesis Ha : µ > µ0 Ha : µ < µ0 Ha : µ 6= µ0 Rejection Region for Level α Test t ≥ tα,n−1 (upper-tailed test) t ≤ −tα,n−1 (lower-tailed test) t ≥ tα/2,n−1 or t ≤ −tα/2,n−1 (two-tailed test)