Activity Number:
|
252
- SPEED: Nonparametrics and Imaging
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 31, 2017 : 3:05 PM to 3:50 PM
|
Sponsor:
|
Section on Nonparametric Statistics
|
Abstract #325215
|
|
Title:
|
Instrument Assisted Partial-Linear Single Index Regression for Errors in Variables Models with Binary Response
|
Author(s):
|
Qianqian Wang* and Yanyuan Ma and Guangren Yang and Xia Cui
|
Companies:
|
University of South Carolina and Penn State University and Jinan University and Zhongshan University
|
Keywords:
|
binary response ;
errors in variables ;
instrumental variables ;
single index model ;
semiparametric efficiency ;
measurement error
|
Abstract:
|
We study partial-linear single index model with error-in-variables when response is binary and instrumental variables are provided. The procedure has two stages. In the first stage, we use instrumental variable method to study the relationship between the observed variable and the instrumental variable. The second stage is to estimate the link function and the parameters in single index by using semi-parametric approach and local linear method. The asymptotic normality is established for both parametric components. An application to real data is illustrated.
|
Authors who are presenting talks have a * after their name.