This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.
Abstract Details
|
Activity Number:
|
618
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 5, 2010 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Health Policy Statistics Section
|
| Abstract - #308946 |
|
Title:
|
Some Sources of Bias and Instability in Binary Imputation by Dichotomization of Model-Based Predicted Probabilities: A Simulation-Based Study
|
|
Author(s):
|
Peter Richardson*+
|
|
Companies:
|
Houston VA Medical Center/HSR&D
|
|
Address:
|
, Houston, TX, ,
|
|
Keywords:
|
imputation ;
epidemiologic studies ;
ROC analysis ;
diagnostic medicine
|
|
Abstract:
|
Common use is made in epidemiological observational studies of imputing missing binary classifications or other indicators by regression modeling followed by dichotomization of the resulting predicted probabilities. In many cases, these models are highly dependent upon categorical predictors. Such practice is prone to dangers such as bias and instability in determination of dichotomization thresholds even when a rigorous decision theory-based approach is taken. By the use of simulations, we explore the magnitude of such bias and instability due to such things as sample size, distribution of model predictors and model mis-specification.
|
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 2010 program
|
2010 JSM Online Program Home
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.