|
Activity Number:
|
121
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
IMS
|
| Abstract - #304953 |
|
Title:
|
Bias Reduction in Generalized Nonlinear Models
|
|
Author(s):
|
Ioannis Kosmidis*+ and David Firth
|
|
Companies:
|
University of Warwick and University of Warwick
|
|
Address:
|
D0.05, Department of Statistics, , Coventry, CV4 7AL, United Kingdom
|
|
Keywords:
|
asymptotic bias correction ; penalized likelihood ; generalized nonlinear models ; pseudo-data
|
|
Abstract:
|
In Firth (1993, \emph{Biometrika}) it was shown how the leading term in the asymptotic bias of the maximum likelihood estimator is removed by adjusting the score vector. In the current work, a more general family of bias-reducing adjustments is developed, for a broad class of generalized nonlinear models. The resulting formulae for the adjusted score vector are computationally convenient, and they directly suggest implementation through an iterative scheme of data adjustment. For generalized linear models a necessary and sufficient condition is given for the existence of a penalized likelihood interpretation of the method along with the form of the resultant penalized likelihoods. An illustrative application to the Goodman row-column association model illustrates the computational simplicity and statistical benefits of bias reduction.
|
- 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 2009 program |
