JSM 2004 - Toronto

Abstract #300806

This is the preliminary program for the 2004 Joint Statistical Meetings in Toronto, Canada. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2004); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

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 2004 Program page
Activity Number: 26
Type: Contributed
Date/Time: Sunday, August 8, 2004 : 2:00 PM to 3:50 PM
Sponsor: Section on Survey Research Methods
Abstract - #300806
Title: M-quantile Models for Small-area Estimation
Author(s): Raymond Chambers*+ and Nikolaos Tzavidis
Companies: University of Southampton and University of Southampton
Address: S3RI, Southampton, International, SO17 1BJ, United Kingdom
Keywords: mixed-effects models ; weighted least squares ; robust inference ; quantile regression ; influence functions
Abstract:

Traditionally, model-based small-area estimation relies on mixed-effects (multilevel) models. Here, we investigate the use of M-quantile models for this purpose. Unlike the mixed-effects approach, which models the expected value of the conditional distribution of the response given the covariates, the M-quantile approach models any set of M-quantile (percentile-like) values of this conditional distribution. By doing so, we avoid normality assumptions on the error term as well as imposing modeling restrictions analogous to random intercepts or random slopes. Instead, between-area variability is captured via variation in area-specific M-quantile "scores." The M-quantile approach is illustrated and contrasted with the mixed effects approach using real-life datasets. Results from Monte Carlo simulation studies indicate that M-quantile models may provide an alternative, and in some cases preferable, solution for small-area estimation problems.

  • 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 2004 program

JSM 2004 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 March 2004