Longitudinal data analysis using nonparametric regression model
This paper proposes nonparametric regression model to analyze longitudinal data. We combine natural cubic spline with generalized estimating equations (GEE) to handle unknown function of the covariate and accounting for the correlation within subjects. Specific condition in which we assume independ...
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2009
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/11363/ http://psasir.upm.edu.my/id/eprint/11363/1/ID%2011363.doc |
| Summary: | This paper proposes nonparametric regression model to analyze longitudinal data. We combine natural cubic spline with generalized estimating equations (GEE) to handle unknown function of the covariate and accounting for the correlation within subjects. Specific condition in which we assume independence, AR-1 and exchangeable correlation structures from each subject with varying sample size are used in the simulation study to assess the efficiency of the estimators. A real data application of the proposed model is illustrated with comparison to parametric model and GEE-smoothing spline under independence assumption. |
|---|