Back To Index Previous Article Next Article Full Text

Statistica Sinica 27 (2017), 1281-1298

GENERALIZED PARTIAL LINEAR MODEL WITH
UNKNOWN LINK AND UNKNOWN BASELINE
FUNCTIONS FOR LONGITUDINAL DATA
Huazhen Lin1, Ling Zhou1 and Binhuan Wang2
1Southwestern University of Finance and Economics and 2New York University

Abstract: In this paper we develop a generalized partial linear model for longitudinal data. In the model, we allow the link and baseline functions to be unknown. We explicitly express the estimators of regression parameters and the baseline function; hence, the computation and programming of our estimators are simple. We show that the proposed estimators of regression parameters and the baseline function are asymptotically normal with a simple variance estimator for the baseline function. In simulation studies, we demonstrate that the proposed nonparametric method is robust with limited loss of efficiency.

Key words and phrases: Generalized partial linear models, kernel method, longitu-dinal data, unknown baseline function, unknown link function.

Back To Index Previous Article Next Article Full Text