Abstract #302119

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); 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 2003 Program page
JSM 2003 Abstract #302119
Activity Number: 450
Type: Contributed
Date/Time: Thursday, August 7, 2003 : 8:30 AM to 10:20 AM
Sponsor: Biometrics Section
Abstract - #302119
Title: Addressing Skewed Data in Multivariate Incomplete Data Settings using Model-Based Multiple Imputation Methods
Author(s): Juwon Song*+ and Thomas R. Belin
Companies: University of California, Los Angeles and University of California, Los Angeles
Address: 10920 Wilshire Blvd., Suite 300, Los Angeles, CA, 90024-6523,
Keywords: incomplete data ; multiple imputation ; skewed distribution
Abstract:

Variables in survey research are often distributed non-normally. In particular, variables representing counts of occurrences are typically skewed, and transformations may not succeed in achieving even approximate normality for such variables. For example, in a study of late-life depression treatment considered by Tang, Belin, and Song (2002), follow-up measures of the number of doctor visits during the previous three months feature many zero responses and a decreasing pattern in the frequency of reports of a positive number of visits. Missing data sometimes arise on these variables, presenting challenges for parametric multiple imputation methods based on an assumption of multivariate normality. Tang, Belin, and Song (2002) reported substantial bias and less-than-nominal coverage in multiple imputation based on the multivariate normal distribution assumption. Here, we explore further the connection between skewness in data distributions and statistical properties of estimates using variations on multivariate normal multiple imputation to handle incomplete data, and we discuss alternative approaches to avoid bias.

  • 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 2003 program

JSM 2003 For information, contact meetings@amstat.org or phone (703) 684-1221. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2003