Activity Number:
|
382
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 15, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Nonparametric Statistics*
|
Abstract - #301403 |
Title:
|
Inference with Estimating Equations. Application to Design-based Inference
|
Author(s):
|
Ioana Schiopu-Kratina*+ and Raluca Balan and Peter Daffer
|
Affiliation(s):
|
Statistics Canada and Université de Sherbrooke and System Planning & Analysis
|
Address:
|
R.H. Coats Building, 16th Floor, Ottawa, Ontario, K1A 0T6, Canada
|
Keywords:
|
estimating equations ; strong and weak convergence ; design-based inference
|
Abstract:
|
This article presents general results on the existence, consistency, and asymptotic normality of estimators obtained as roots of very general estimating equations (EE). We consider conditions for strong and weak convergence of such estimators on arbitrary probability spaces, and we generalize results of K-H.--Yuan and R.I. Jennrich (JMA 65, 1998) and K. Chen et al. ( Ann. Stat. 27, 1999). The results are then applied to GEE. We also consider inference in finite populations with probabilities induced by the survey design.
|