Activity Number:
|
369
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 2, 2016 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Nonparametric Statistics
|
Abstract #318983
|
View Presentation
|
Title:
|
Semiparametric Efficient Estimators in Heteroscedastic Error Models
|
Author(s):
|
Mijeong Kim* and Yanyuan Ma
|
Companies:
|
Ewha Womans University and University of South Carolina
|
Keywords:
|
Semiparametric methods ;
heteroscedastic error models ;
efficiency bounds
|
Abstract:
|
We consider several frequently encounted regression models where the mean and variance functions are specified up to some parameters. An important point that we would like to emphasize through the series of analyses is that different assump- tions on the standardized regression errors yield very different efficiency bounds of the corresponding estimators. Consequently, all aspects of the assumptions need to be taken into account specifically in constructing their corresponding efficient estimators. Our study clarifies the relation between the regression error assumptions and their respectively efficiency bounds, under the general framework of regression with heteroscedastic errors.
|
Authors who are presenting talks have a * after their name.