A Truncated Spline and Local Linear Mixed Estimator in Nonparametric Regression for Longitudinal Data and Its Application

Longitudinal data modeling is widely carried out using parametric methods. However, when the model misspecified, obtained estimator might be severely biased and lead to erroneous conclusions. In this study, we propose a new estimation method for longitudinal mixed in nonparametric regression. The objective of study was estimate regression curve two combined estimators: truncated spline local linear. weighted least square with two-stage procedure used obtain proposed model. To account within-subject correlations data, symmetric weight matrix given estimation. best determined by minimizing generalized cross-validation value. Furthermore, an application dataset poverty gap index Bengkulu Province, Indonesia, conducted illustrate performance estimator. Compared single estimator, linear had better based on GCV Additionally, empirical results indicated that could explain variation exceptionally well.