|
Activity Number:
|
114
- The Need and Methods for Routine Inclusion of Model Uncertainty in Statistical Results
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 31, 2017 : 8:30 AM to 10:20 PM
|
|
Sponsor:
|
Quality and Productivity Section
|
|
Abstract #322752
|
|
|
Title:
|
Comparison of Model selction and Averaging for Collaborative Studies
|
|
Author(s):
|
Andrew Rukhin*
|
|
Companies:
|
|
|
Keywords:
|
common mean ;
likelihood ratio test ;
information criterion ;
heteroscedasticity ;
model selection ;
restricted likelihood
|
|
Abstract:
|
There are many collaborative studies where the reported within-study uncertainty estimates are unreliable but can be considered as lower bounds to the true uncertainties. Motivated by such examples this talk provides a method to choose a statistical model or an average of several models for heterogeneous observations with unknown variances which allow for positive lower bounds. The assessment of the final uncertainty of the derived estimators is also discussed.
|
Authors who are presenting talks have a * after their name.
