Activity Number:
|
554
- Novel Methods in Longitudinal Data Analysis
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 11, 2022 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biometrics Section
|
Abstract #322191
|
|
Title:
|
Unified Inferences for Sparse and Dense Longitudinal Models with Irregular Time AR Error Process
|
Author(s):
|
Yan Fang* and Yin Hang and Jinhong You
|
Companies:
|
Shanghai University of International Business and Economics and Guangdong University of Finance and Shanghai University of Finance and Economics
|
Keywords:
|
Longitudinal Data;
Nonparametric Method;
Unified Weighted Local Linear Estimate;
Irregular Time AR Error Process;
Unified Inference
|
Abstract:
|
We consider nonparametric estimation of the mean for longitudinal data. We propose an irregular time AR model for the error process and develop a unified two-stage estimator based on the weighted local linear approach for the mean. Under certain regular conditions, we establish both the asymptotic normality and the almost sure rates of convergence of the proposed nonparametric estimator. In addition, we derive the asymptotic bias and variance of the estimator. An appealing feature of our proposed method is that a unified inference can be done without the need of making distinction whether the data are sparse or dense. The effectiveness of the proposed unified inference is demonstrated through a few simulation studies and an analysis of ACT-315 data.
|
Authors who are presenting talks have a * after their name.