This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Outline Introduction to Confidence Intervals Nonparametric CIs for Means and Proportions The issue of Precision Nonparametric CIs for the Median Lecture 23 Chapter 8: Confidence and Prediction Intervals Michael Akritas Michael Akritas Lecture 23 Chapter 8: Confidence and Prediction Intervals Outline Introduction to Confidence Intervals Nonparametric CIs for Means and Proportions The issue of Precision Nonparametric CIs for the Median Introduction to Confidence Intervals Nonparametric CIs for Means and Proportions Nonparametric CIs for Means Nonparametric CIs for Proportions The issue of Precision Generalities Sample size determination for improved precision: μ Sample size determination for improved precision: p Nonparametric CIs for the Median Michael Akritas Lecture 23 Chapter 8: Confidence and Prediction Intervals Outline Introduction to Confidence Intervals Nonparametric CIs for Means and Proportions The issue of Precision Nonparametric CIs for the Median Bounding the Error of Estimation Michael Akritas Lecture 23 Chapter 8: Confidence and Prediction Intervals Outline Introduction to Confidence Intervals Nonparametric CIs for Means and Proportions The issue of Precision Nonparametric CIs for the Median Bounding the Error of Estimation I By the CLT, if n is large, most estimators, b θ , are approximately normally distributed. Michael Akritas Lecture 23 Chapter 8: Confidence and Prediction Intervals Outline Introduction to Confidence Intervals Nonparametric CIs for Means and Proportions The issue of Precision Nonparametric CIs for the Median Bounding the Error of Estimation I By the CLT, if n is large, most estimators, b θ , are approximately normally distributed. I For example, the estimators of Chapter 6, and the moment estimators, are all approximately normal for large n . Michael Akritas Lecture 23 Chapter 8: Confidence and Prediction Intervals Outline Introduction to Confidence Intervals Nonparametric CIs for Means and Proportions The issue of Precision Nonparametric CIs for the Median Bounding the Error of Estimation I By the CLT, if n is large, most estimators, b θ , are approximately normally distributed. I For example, the estimators of Chapter 6, and the moment estimators, are all approximately normal for large n . I These estimators are either unbiased or nearly unbiased, and their estimated standard error, b σ ˆ θ , provides a reliable estimate of σ ˆ θ . Thus, if θ is the true value of the parameter, ˆ θ · ∼ N θ, b σ 2 ˆ θ . Michael Akritas Lecture 23 Chapter 8: Confidence and Prediction Intervals Outline Introduction to Confidence Intervals Nonparametric CIs for Means and Proportions The issue of Precision Nonparametric CIs for the Median Bounding the Error of Estimation I By the CLT, if n is large, most estimators, b θ , are approximately normally distributed....
View
Full
Document
This note was uploaded on 03/19/2009 for the course STAT 401 taught by Professor Akritas during the Spring '00 term at Penn State.
 Spring '00
 Akritas

Click to edit the document details