3710ProblemCase1 - n x z-= Step 3 Since the H A here is of...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
MSCI 3710 LECTURE NOTES Dr. Nick Evangelopoulos Hypothesis Testing Problem Case 1 Situation: There is an unknown population mean μ . We know the sample mean x and the sample size n . The population standard deviation σ is given. Then, we do a z test for one mean. (Alternatively, sigma is not given, only the sample st. deviation s is given, but we have a large sample n > 30; then we can do an approximate z test .) Example: Refer to the Everglo example, 2005 text, p.311. We suspect might not be 400 as claimed. A sample of 100 had x = 411. The population st. deviation is known to be σ = 42.5. Test at a = 0.10. Solution: Step1 H 0 : = 400 H A : ≠ 400 Step 2 Given that H 0 is true (which is our assumption anyway), the following quantity is expected to have a standard normal distribution:
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: n x z-= Step 3 Since the H A here is of the not-equal-to type, the test is two-tailed. Each tail has a probability equal to a /2. Using a z table we find the critical value to be 1.645. The Decision Rule is Reject H if the observed z value is more extreme than the critical z value. Step 4 The calculated value is: 100 5 . 42 400 411 *-= z = (411 400) / (42.5 / SQRT(100)) = 2.59. Step 5 Conclusion: Reject the null hypothesis H since 2.59 > 1.645, i.e., the calculated value is more extreme than the critical value(s). Therefore, the average lifetime of a light bulb is not equal to 400. a /2 z cr a /2 z cr 0.05 1.645 0.05 1.645 z cr z cr z * = 2.59...
View Full Document

This note was uploaded on 03/07/2011 for the course DSCI 3710 taught by Professor Dr.jay during the Spring '08 term at North Texas.

Ask a homework question - tutors are online