JSM Preliminary Online Program
This is the preliminary program for the 2006 Joint Statistical Meetings in Seattle, Washington.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2006 Program page

Activity Number: 132
Type: Topic Contributed
Date/Time: Monday, August 7, 2006 : 10:30 AM to 12:20 PM
Sponsor: Section on Health Policy Statistics
Abstract - #305850
Title: Confidence Intervals on General Variance Components Model: Modified Large Sample Approach
Author(s): Yonghee Lee*+
Companies: Ewha Womans University
Address: 111 Daehyundong SedaemunGu, Seoul, 120-750, South Korea
Keywords: variance components ; confidence intervals ; unbalanced designs
Abstract:

Inference on variance components is not straight forward because of the complex sampling distributions of ANOVA-type estimators. Especially in unbalanced designs, it is more difficult to make inferences on the function of variance components, as the sum of squares from ANOVA may not follow chi-square distribution or sum of squares are not independent. By using the results of Lee, Shao, and Chow (2004), I will introduce an extension of Modified Large Sample (MLS) method for construction of confidence intervals on a linear combination of variance components when ANOVA sum of squares do not follow a chi-square distribution or when estimators are dependent. The proposed method is based on an approximation to a linear combination of chi-square random variables whose degrees of freedom are one and bootstrap calibration. I will discuss applications to nested unbalanced designs as well.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2006 program

JSM 2006 For information, contact jsm@amstat.org or phone (888) 231-3473. If you have questions about the Continuing Education program, please contact the Education Department.
Revised April, 2006