JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 132
Type: Contributed
Date/Time: Monday, August 10, 2015 : 8:30 AM to 10:20 AM
Sponsor: Korean International Statistical Society
Abstract #314968
Title: A Transformation-Based Approach to Sample Size Calculation for Logistic Regression
Author(s): Seongho Kim* and Elisabeth Heath and Lance Heilbrun
Companies: Wayne State University/Karmanos Cancer Institute and Wayne State University/Karmanos Cancer Institute and Wayne State University/Karmanos Cancer Institute
Keywords: Logistic regression ; Logit-normal distribution ; Power calculation ; Sample size determination ; Transformation
Abstract:

The sample size determination for multiple logistic regressions requires additional information, the coefficient of determination of a covariate of interest with other covariates, which is often unavailable in practice, even though the sample size for simple logistic regression can be readily calculated using existing methods. The response variable of logistic regression follows a logit-normal distribution that can be generated from a logistic transformation of a normal distribution. Using a normal transformation of outcome measures, we propose new methods of the sample size calculation for simple and multiple logistic regressions. Our simulation studies show several advantages of the developed approach over the currently available methods: much smaller required sample size, no need of the coefficient of determination for multiple logistic regression, and available interim or group-sequential designs.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home