Abstract Details
Activity Number:
|
232
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 4, 2014 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biometrics Section
|
Abstract #313242
|
View Presentation
|
Title:
|
The Role of Akaike Weights in Model Selection
|
Author(s):
|
Jessica K. Kohlschmidt*+ and Kati S. Maharry and Clara D. Bloomfield
|
Companies:
|
Ohio State University and Ohio State University and Ohio State University
|
Keywords:
|
Modeling ;
AIC ;
Akaike weights ;
Oncology
|
Abstract:
|
When investigating the usefulness of a molecular marker, multivariable models are often considered. These models are used to analyze the influence of a marker on outcome while considering variables that have been shown to be important. There are often several models that could be used to describe outcome using the variables that have been collected. The question becomes which model provides the best insight into the data. In this talk, we will look at a number of models that could be used for a particular analysis and then using Akaike weights make a determination as to the model that is most likely to be the "best". This is a statistical approach to modeling instead of choosing a model based on perceived biological relationships. The approach using Akaike weights considers the possible models and uses the cumulative information to make conclusions regarding the feasibility of each of the models in light of the data available. We provide this as an alternative approach to presenting models in oncology research. As we have more data available to researchers on a fixed number of patients (i.e. massive array data) we are in need of solutions such as this to aid in research decisions.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please 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.
Copyright © American Statistical Association.