Abstract #302360

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 #302360
Activity Number: 104
Type: Invited
Date/Time: Monday, August 4, 2003 : 10:30 AM to 12:20 PM
Sponsor: Biometrics Section
Abstract - #302360
Title: Exact Target Estimation in Logistic Regression
Author(s): Nitin Patel*+ and Pralay Senchaudhuri and Brent Coull
Companies: Cytel Software Corporation and Cytel Software Corporation and Harvard School of Public Health
Address: 675 Massachusetts Ave., Cambridge, MA, 02139-3309,
Keywords:
Abstract:

Recent research in small sample methods for categorical data has resulted in a variety of methods for conducting exact hypothesis testing and power calculations. In contrast, relatively little work has focused on point estimation, even though it is well-known that maximum likelihood estimates of coefficients in generalized linear models are biased in small samples. One approach to bias and variance reduction in logistic regression models is target estimation, which takes as an estimate of a parameter vector the value that sets the expected value of the maximum likelihood estimator equal to the observed value. We propose a network algorithm approach for target estimation of regression parameters in logistic regression models. Advantages of this approach include the exact calculation of the bias function for maximum likelihood estimates, the ability to reduce the dimension of this bias function via conditioning, straightforward standard error calculations, and the availability of existing software to perform the calculations.

  • 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