Liu estimator in partly linear regression models with correlated errors

ABSTRACT

This article is concerned with the parameter estimation in partly linear regression models when the errors are dependent. To overcome the multicollinearity problem, a generalized Liu estimator is proposed. The theoretical properties of the proposed estimator and its relationship with some existing methods designed for partly linear models are investigated. Finally, a hypothetical data is conducted to illustrate some of the theoretical results.

KEYWORDS:

• Least squares
• Liu estimator
• Partly linear models
• Ridge estimator
• Smoothing parameter

MATHEMATICS SUBJECT CLASSIFICATION:

• Primary 62G08
• Secondary 62J05
• 62J07

Acknowledgment

The authors are thankful to the editor and referee for their constructive and valuable suggestions and comments that have improved the presentation of the paper greatly.