This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Study Questions These are examples of the kinds of questions that might appear on the final exam. You might also review the Conceptual Exercises in the Sleuth ; “thought” questions in the homework assignments; the Glossary; and announcements and comments appearing on Blackboard. 1. Suppose we measure some response in samples from three populations, each with the same underlying variance, σ 2 . To test the null hypothesis that γ 1 ≡ μ 1- 1 2 ( μ 2 + μ 3 ) = 0 , we use the statistic ˆ γ 1 SE (ˆ γ 1 ) = ¯ Y 1- 1 2 ( ¯ Y 2 + ¯ Y 3 ) radicalBig MSE [1 /n 1 + (1 / 4) · (1 /n 2 + 1 /n 3 )] . If γ 2 ≡ 2 μ 1- ( μ 2 + μ 3 ) , show that the statistic for testing the null hypothesis γ 2 = 0 is identical to the above statistic. 2. What is wrong with this formulation of the regression model: μ ( Y | X ) = β + β 1 X + ǫ ? 3. In the Spock example, write down a null hypothesis of possible interest that involves more than two populations (judges). Explain what question you would be addressing by testingthan two populations (judges)....
View Full Document
This note was uploaded on 11/17/2009 for the course STAT 511 taught by Professor Paulmurtaugh during the Fall '09 term at Oregon State.
- Fall '09