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: Summary of the Risk in Decision Making
Actual Situation Statistical Decision
Do not reject H0 H0 True Correct decision Probability = 1 – α, the confidence level Type I error Probability = α, the significance level H0 False Type II error Probability = β Correct decision Probability = 1 – β , the power Reject H0 Summary of Significance Tests for the Mean of a Normal Population
A. n large: σ known *
Null Hypothesis, H0 µ = µ 0 or µ ≤ µ 0 Alternative Hypothesis, Ha or H1 µ > µ0 Test Statistic Decision Rule Reject H0 if z > zα Reject H0 if z < –zα z= z= z= x − µ0 µ = µ 0 or µ ≥µ 0 µ = µ0 µ < µ0 µ ≠µ 0 σn x − µ0 σn x − µ0 σ
n z > zα 2 Reject H0 if or z < − zα 2 * If σ unknown, use s to estimate σ. B. n small: normal population and σ unknown
Null Hypothesis, H0 µ = µ 0 or µ ≤ µ 0 Alternative Hypothesis, Ha or H1 µ > µ0 Test Statistic Decision Rule Reject H0 if t > tn1,a t= t= t= x − µ0 sn x − µ0 sn x − µ0 s n µ = µ 0 or µ ≥µ 0 µ = µ0 µ < µ0 µ ≠µ 0 Reject H0 if t < –tn1,α t > t n − 1, α 2 Reject H0 if or t < −t n −1,α 2 © 1999 by Harvey A. Singer © 1999 by Harvey A. Singer ...
View Full
Document
 Fall '08
 SINGER

Click to edit the document details