Abstract
Tabakan and Akdeniz introduced a difference-based ridge estimator in partially linear model. In this article a generalized difference-based weighted mixed ridge estimator in partially linear model is presented by combining the ideas underlying the generalized difference-based weighted mixed estimator and the generalized difference-based ridge estimator, when stochastic linear restrictions are assumed to hold. We also discussed the properties of the new estimator and compared the new estimator to the difference-based ridge estimator and difference-based weighted mixed estimator in the mean squared error matrix. A method to select the biasing parameters is also discussed. Finally, a numerical example and a simulation study are given to show the performances of the estimators.
Acknowledgements
The authors are highly obliged to the editor and the reviewer for the comments and suggestions which improved the paper in its present form.